Multi-Constituent Analysis with All Cohorts - Post-Publication Time-Weighted Analysis

1 Simmons - Importing Datasets

Here I am importing the file which contains monthly PM level estimates by satellite at nearest lon/lat to Simmons patient residential addresses. These are linked to the patient ID.

outfile1 <- here("Simmons_fILD_2000_2018_PM25_2021_09_08.xlsx")
PM <- read_excel(outfile1)

Here I am importing the file which I used for my ADI work that contains the baseline clinical and demographic data for 1425 patients who have ADI

outfile2 <- here("Simmons_fILDPts_BaselineData_2022_03_28.xlsx")
Simm <- read_excel(outfile2)

2 Simmons - Simplifying PM Dataframe

PM <- PM %>% dplyr::select(!c(nrow, dist, lon, lat))
colnames(PM)

##   [1] "ID"         "PM25_jan00" "PM25_feb00" "PM25_mar00" "PM25_apr00"
##   [6] "PM25_may00" "PM25_jun00" "PM25_jul00" "PM25_aug00" "PM25_sep00"
##  [11] "PM25_oct00" "PM25_nov00" "PM25_dec00" "PM25_jan01" "PM25_feb01"
##  [16] "PM25_mar01" "PM25_apr01" "PM25_may01" "PM25_jun01" "PM25_jul01"
##  [21] "PM25_aug01" "PM25_sep01" "PM25_oct01" "PM25_nov01" "PM25_dec01"
##  [26] "PM25_jan02" "PM25_feb02" "PM25_mar02" "PM25_apr02" "PM25_may02"
##  [31] "PM25_jun02" "PM25_jul02" "PM25_aug02" "PM25_sep02" "PM25_oct02"
##  [36] "PM25_nov02" "PM25_dec02" "PM25_jan03" "PM25_feb03" "PM25_mar03"
##  [41] "PM25_apr03" "PM25_may03" "PM25_jun03" "PM25_jul03" "PM25_aug03"
##  [46] "PM25_sep03" "PM25_oct03" "PM25_nov03" "PM25_dec03" "PM25_jan04"
##  [51] "PM25_feb04" "PM25_mar04" "PM25_apr04" "PM25_may04" "PM25_jun04"
##  [56] "PM25_jul04" "PM25_aug04" "PM25_sep04" "PM25_oct04" "PM25_nov04"
##  [61] "PM25_dec04" "PM25_jan05" "PM25_feb05" "PM25_mar05" "PM25_apr05"
##  [66] "PM25_may05" "PM25_jun05" "PM25_jul05" "PM25_aug05" "PM25_sep05"
##  [71] "PM25_oct05" "PM25_nov05" "PM25_dec05" "PM25_jan06" "PM25_feb06"
##  [76] "PM25_mar06" "PM25_apr06" "PM25_may06" "PM25_jun06" "PM25_jul06"
##  [81] "PM25_aug06" "PM25_sep06" "PM25_oct06" "PM25_nov06" "PM25_dec06"
##  [86] "PM25_jan07" "PM25_feb07" "PM25_mar07" "PM25_apr07" "PM25_may07"
##  [91] "PM25_jun07" "PM25_jul07" "PM25_aug07" "PM25_sep07" "PM25_oct07"
##  [96] "PM25_nov07" "PM25_dec07" "PM25_jan08" "PM25_feb08" "PM25_mar08"
## [101] "PM25_apr08" "PM25_may08" "PM25_jun08" "PM25_jul08" "PM25_aug08"
## [106] "PM25_sep08" "PM25_oct08" "PM25_nov08" "PM25_dec08" "PM25_jan09"
## [111] "PM25_feb09" "PM25_mar09" "PM25_apr09" "PM25_may09" "PM25_jun09"
## [116] "PM25_jul09" "PM25_aug09" "PM25_sep09" "PM25_oct09" "PM25_nov09"
## [121] "PM25_dec09" "PM25_jan10" "PM25_feb10" "PM25_mar10" "PM25_apr10"
## [126] "PM25_may10" "PM25_jun10" "PM25_jul10" "PM25_aug10" "PM25_sep10"
## [131] "PM25_oct10" "PM25_nov10" "PM25_dec10" "PM25_jan11" "PM25_feb11"
## [136] "PM25_mar11" "PM25_apr11" "PM25_may11" "PM25_jun11" "PM25_jul11"
## [141] "PM25_aug11" "PM25_sep11" "PM25_oct11" "PM25_nov11" "PM25_dec11"
## [146] "PM25_jan12" "PM25_feb12" "PM25_mar12" "PM25_apr12" "PM25_may12"
## [151] "PM25_jun12" "PM25_jul12" "PM25_aug12" "PM25_sep12" "PM25_oct12"
## [156] "PM25_nov12" "PM25_dec12" "PM25_jan13" "PM25_feb13" "PM25_mar13"
## [161] "PM25_apr13" "PM25_may13" "PM25_jun13" "PM25_jul13" "PM25_aug13"
## [166] "PM25_sep13" "PM25_oct13" "PM25_nov13" "PM25_dec13" "PM25_jan14"
## [171] "PM25_feb14" "PM25_mar14" "PM25_apr14" "PM25_may14" "PM25_jun14"
## [176] "PM25_jul14" "PM25_aug14" "PM25_sep14" "PM25_oct14" "PM25_nov14"
## [181] "PM25_dec14" "PM25_jan15" "PM25_feb15" "PM25_mar15" "PM25_apr15"
## [186] "PM25_may15" "PM25_jun15" "PM25_jul15" "PM25_aug15" "PM25_sep15"
## [191] "PM25_oct15" "PM25_nov15" "PM25_dec15" "PM25_jan16" "PM25_feb16"
## [196] "PM25_mar16" "PM25_apr16" "PM25_may16" "PM25_jun16" "PM25_jul16"
## [201] "PM25_aug16" "PM25_sep16" "PM25_oct16" "PM25_nov16" "PM25_dec16"
## [206] "PM25_jan17" "PM25_feb17" "PM25_mar17" "PM25_apr17" "PM25_may17"
## [211] "PM25_jun17" "PM25_jul17" "PM25_aug17" "PM25_sep17" "PM25_oct17"
## [216] "PM25_nov17" "PM25_dec17" "PM25_jan18" "PM25_feb18" "PM25_mar18"
## [221] "PM25_apr18" "PM25_may18" "PM25_jun18" "PM25_aug18" "PM25_sep18"
## [226] "PM25_oct18" "PM25_nov18" "PM25_dec18"

PM <- PM %>% 
  pivot_longer(cols=c(2:228), names_to="PM_date", names_prefix="PM25_")

PMx <- PM 
PMx$PM_date <- gsub("jan", "01-01-20", PMx$PM_date)
PMx$PM_date <- gsub("feb", "01-02-20", PMx$PM_date)
PMx$PM_date <- gsub("mar", "01-03-20", PMx$PM_date)
PMx$PM_date <- gsub("apr", "01-04-20", PMx$PM_date)
PMx$PM_date <- gsub("may", "01-05-20", PMx$PM_date)
PMx$PM_date <- gsub("jun", "01-06-20", PMx$PM_date)
PMx$PM_date <- gsub("jul", "01-07-20", PMx$PM_date)
PMx$PM_date <- gsub("aug", "01-08-20", PMx$PM_date)
PMx$PM_date <- gsub("sep", "01-09-20", PMx$PM_date)
PMx$PM_date <- gsub("oct", "01-10-20", PMx$PM_date)
PMx$PM_date <- gsub("nov", "01-11-20", PMx$PM_date)
PMx$PM_date <- gsub("dec", "01-12-20", PMx$PM_date)

PMx$PM_date <- format(as.Date(PMx$PM_date, format="%d-%m-%Y"),"%Y-%m-%d")
PMx$PM_date <- as.Date(PMx$PM_date)
PM <- PMx
rm(PMx)

3 Simmons - Simplifying Simm Dataframe

3.1 Simmons - Death/Transplant/Censoring Date

Extracting year of diagnosis and year of death/transplant/censoring

#Start with the year of diagnosis
Simm <- Simm %>% 
  mutate(dx_yrmo = format(as.Date(Simm$dx_date, format="%Y-%m-%d"),"%Y-%m"))
Simm <- Simm %>% 
  mutate(dx_yr = format(as.Date(Simm$dx_date, format="%Y-%m-%d"),"%Y"))
Simm$dx_yr <- as.numeric(Simm$dx_yr)

#Then the year of death or lung transplant
Simm <- Simm %>% 
  mutate(deathORtx_date = if_else(!is.na(tx_date), tx_date, death_date))
Simm <- Simm %>% 
  mutate(deathORtx_yrmo = format(as.Date(Simm$deathORtx_date, format="%Y-%m-%d"),"%Y-%m"))

#Then the year the records were last updated (i.e. year of censoring)
Simm <- Simm %>% 
  mutate(DeathTxCensor_date = if_else(!is.na(deathORtx_date), deathORtx_date, last_updated))
Simm <- Simm %>% 
  mutate(censor_yrmo = format(as.Date(Simm$DeathTxCensor_date, format="%Y-%m-%d"),"%Y-%m"))

3.2 Simmons - Removing Unnecessary Columns and Correcting Dates

Simm <- Simm %>% dplyr::select(!c(ADI_state, lat, lon, City, State, zip5, UPMC_lastvisit, Simmons_lastvisit, pkyrs, fev1_pre, fev1_pct, fvc_pre, dlco_pre, pft_timefromdx, ethnicity, dx_type, deathORtx_date, deathORtx_yrmo))
Simm <- Simm %>% 
  mutate_at(c("dob","death_date", "last_updated", "tx_date", "dx_date", "consent_date", "pft_date", "DeathTxCensor_date"), as.Date)
str(Simm)

## tibble [1,425 × 32] (S3: tbl_df/tbl/data.frame)
##  $ ID                : num [1:1425] 1097 1405 2796 4742 7898 ...
##  $ dob               : Date[1:1425], format: "1937-08-15" "1956-07-21" ...
##  $ ADI_nat           : num [1:1425] 67 73 68 62 21 34 25 23 68 NA ...
##  $ death_date        : Date[1:1425], format: "2008-12-09" NA ...
##  $ last_updated      : Date[1:1425], format: "2021-01-27" "2021-01-27" ...
##  $ tx_date           : Date[1:1425], format: NA "2019-04-18" ...
##  $ dx_date           : Date[1:1425], format: "2002-01-24" "1999-07-26" ...
##  $ consent_date      : Date[1:1425], format: "2003-02-27" "2003-10-30" ...
##  $ pft_date          : Date[1:1425], format: "2002-01-18" "1999-07-26" ...
##  $ fvc_pct           : num [1:1425] 31.9 53.6 82.5 47.6 62 ...
##  $ dlco_pct          : num [1:1425] NA 38.1 98.2 65.3 42 ...
##  $ status            : num [1:1425] 1 2 0 0 0 1 1 1 1 0 ...
##  $ age_dx            : num [1:1425] 64.4 44.7 42.3 45.5 45.2 ...
##  $ time_censoring    : num [1:1425] 6.72 19.93 20.43 19.16 1.45 ...
##  $ time_death        : num [1:1425] 6.87 NA NA NA NA ...
##  $ time_tx           : num [1:1425] NA 18 NA NA NA ...
##  $ time_deathORtx    : num [1:1425] 6.87 18.05 NA NA NA ...
##  $ time_DeathTxCensor: num [1:1425] 6.87 18.05 20.43 19.16 1.45 ...
##  $ sex               : chr [1:1425] "F" "F" "M" "F" ...
##  $ race              : chr [1:1425] "B" "W" "W" "W" ...
##  $ died              : chr [1:1425] "1" "0" "0" "0" ...
##  $ txed              : chr [1:1425] "0" "1" "0" "0" ...
##  $ deadORtx          : chr [1:1425] "1" "1" "0" "0" ...
##  $ dx                : chr [1:1425] "SSC_ILD" "SSC_ILD" "RA_ILD" "MCTD" ...
##  $ smokeHx           : chr [1:1425] NA "Former" "Never" NA ...
##  $ dich_Race         : chr [1:1425] "Non-White" "White" "White" "White" ...
##  $ dich_smoking      : chr [1:1425] NA "Ever" "Never" NA ...
##  $ dx_group          : chr [1:1425] "CTD-ILD" "CTD-ILD" "CTD-ILD" "CTD-ILD" ...
##  $ dx_yrmo           : chr [1:1425] "2002-01" "1999-07" "2003-10" "1999-12" ...
##  $ dx_yr             : num [1:1425] 2002 1999 2003 1999 2015 ...
##  $ DeathTxCensor_date: Date[1:1425], format: "2008-12-09" "2019-04-18" ...
##  $ censor_yrmo       : chr [1:1425] "2008-12" "2019-04" "2021-01" "2021-01" ...

3.3 Simmons - Correcting Smoking

Need to correct smoking variables

Simm$smokeHx <- as.character(Simm$smokeHx) 
Simm <- Simm %>% mutate(smokeHx1=if_else(is.na(smokeHx), "Unknown", smokeHx))

#now need to make new dich_smoking category
Simm$dich_smoking <- as.character(Simm$dich_smoking) 
Simm <- Simm %>% mutate(dich_smoking1=if_else(is.na(dich_smoking),"Unknown", dich_smoking))

Now need to remove old smoking variables and rename new ones

Simm <- Simm %>% dplyr::select(-c(smokeHx, dich_smoking))
Simm <- Simm %>% rename(c("smokeHx"="smokeHx1", "dich_smoking"="dich_smoking1"))
Simm$smokeHx <- as.factor(Simm$smokeHx)
Simm$dich_smoking <- as.factor(Simm$dich_smoking)
Simm$smokeHx <- fct_relevel(Simm$smokeHx, c("Never","Former","Always","Unknown"))
Simm$dich_smoking <- fct_relevel(Simm$dich_smoking, c("Never","Ever","Unknown"))

3.4 Simmons - Correcting Factors

Need to correct other factor variables

Simm$sex <- fct_relevel(Simm$sex, c("M","F"))
Simm$race <- fct_relevel(Simm$race, c("W","B","A","N","U"))
Simm$dich_Race <- fct_relevel(Simm$dich_Race, c("White","Non-White"))
Simm$dx <- fct_relevel(Simm$dx, c("IPF"))
Simm$dx_group <- fct_relevel(Simm$dx_group, c("IPF"))
Simm <- Simm %>% mutate_at(c("status","died", "txed", "deadORtx"), as.factor)

3.5 Simmons - Creating New Variables

3.5.1 Simmons - IPF vs Other Diagnosis

Simm <- Simm %>% mutate(dx_IPF=ifelse(dx=="IPF", "IPF", "not_IPF"))  
Simm$dx_IPF <- fct_relevel(Simm$dx_IPF, c("IPF"))

3.5.2 Simmons - Disadvantage Distribution

Creating this empirical cumulative distribution will allow us to combine the analyses of all three cohorts even though the measurements for disadvantage are different between the three.

plot(ecdf(Simm$ADI_nat))

Simm$disadv <- ecdf(Simm$ADI_nat)(Simm$ADI_nat)

4 Simmons - Modifying Simmons Dataset

Take down Simmons dataset to the 1424 patients with complete data (currently at 1425)

IDs <- as.data.table(unique(PM$ID))
IDs <- IDs %>% rename("ID"="V1")
Simm <- left_join(IDs, Simm, by="ID")

Simm <- Simm %>% mutate(days_DeathTxCensor=(time_DeathTxCensor*365.25))

Longest time_DeathTxCensor= 20.427105yrs =7461days

30-day longest time interval would be 7470 days

start <- seq(1, 7441, by = 30)
start

##   [1]    1   31   61   91  121  151  181  211  241  271  301  331  361  391  421
##  [16]  451  481  511  541  571  601  631  661  691  721  751  781  811  841  871
##  [31]  901  931  961  991 1021 1051 1081 1111 1141 1171 1201 1231 1261 1291 1321
##  [46] 1351 1381 1411 1441 1471 1501 1531 1561 1591 1621 1651 1681 1711 1741 1771
##  [61] 1801 1831 1861 1891 1921 1951 1981 2011 2041 2071 2101 2131 2161 2191 2221
##  [76] 2251 2281 2311 2341 2371 2401 2431 2461 2491 2521 2551 2581 2611 2641 2671
##  [91] 2701 2731 2761 2791 2821 2851 2881 2911 2941 2971 3001 3031 3061 3091 3121
## [106] 3151 3181 3211 3241 3271 3301 3331 3361 3391 3421 3451 3481 3511 3541 3571
## [121] 3601 3631 3661 3691 3721 3751 3781 3811 3841 3871 3901 3931 3961 3991 4021
## [136] 4051 4081 4111 4141 4171 4201 4231 4261 4291 4321 4351 4381 4411 4441 4471
## [151] 4501 4531 4561 4591 4621 4651 4681 4711 4741 4771 4801 4831 4861 4891 4921
## [166] 4951 4981 5011 5041 5071 5101 5131 5161 5191 5221 5251 5281 5311 5341 5371
## [181] 5401 5431 5461 5491 5521 5551 5581 5611 5641 5671 5701 5731 5761 5791 5821
## [196] 5851 5881 5911 5941 5971 6001 6031 6061 6091 6121 6151 6181 6211 6241 6271
## [211] 6301 6331 6361 6391 6421 6451 6481 6511 6541 6571 6601 6631 6661 6691 6721
## [226] 6751 6781 6811 6841 6871 6901 6931 6961 6991 7021 7051 7081 7111 7141 7171
## [241] 7201 7231 7261 7291 7321 7351 7381 7411 7441

end <- seq(30, 7470, by = 30)
end

##   [1]   30   60   90  120  150  180  210  240  270  300  330  360  390  420  450
##  [16]  480  510  540  570  600  630  660  690  720  750  780  810  840  870  900
##  [31]  930  960  990 1020 1050 1080 1110 1140 1170 1200 1230 1260 1290 1320 1350
##  [46] 1380 1410 1440 1470 1500 1530 1560 1590 1620 1650 1680 1710 1740 1770 1800
##  [61] 1830 1860 1890 1920 1950 1980 2010 2040 2070 2100 2130 2160 2190 2220 2250
##  [76] 2280 2310 2340 2370 2400 2430 2460 2490 2520 2550 2580 2610 2640 2670 2700
##  [91] 2730 2760 2790 2820 2850 2880 2910 2940 2970 3000 3030 3060 3090 3120 3150
## [106] 3180 3210 3240 3270 3300 3330 3360 3390 3420 3450 3480 3510 3540 3570 3600
## [121] 3630 3660 3690 3720 3750 3780 3810 3840 3870 3900 3930 3960 3990 4020 4050
## [136] 4080 4110 4140 4170 4200 4230 4260 4290 4320 4350 4380 4410 4440 4470 4500
## [151] 4530 4560 4590 4620 4650 4680 4710 4740 4770 4800 4830 4860 4890 4920 4950
## [166] 4980 5010 5040 5070 5100 5130 5160 5190 5220 5250 5280 5310 5340 5370 5400
## [181] 5430 5460 5490 5520 5550 5580 5610 5640 5670 5700 5730 5760 5790 5820 5850
## [196] 5880 5910 5940 5970 6000 6030 6060 6090 6120 6150 6180 6210 6240 6270 6300
## [211] 6330 6360 6390 6420 6450 6480 6510 6540 6570 6600 6630 6660 6690 6720 6750
## [226] 6780 6810 6840 6870 6900 6930 6960 6990 7020 7050 7080 7110 7140 7170 7200
## [241] 7230 7260 7290 7320 7350 7380 7410 7440 7470

Repeat the list of intervals 1424 times (number of patients in Simmons)

start <- rep(start, times=1424)
end <- rep(end, times=1424)
intervals <- as.data.frame(cbind(start, end))

Add ID column to intervals

IDs <- rep(Simm$ID, each=249)
intervals <- as.data.frame(cbind(IDs, intervals))
intervals <- intervals %>% rename("ID"="IDs")

Join Simm and intervals

Simm <- left_join(intervals, Simm, by="ID")

Determine if event occurred during interval

Simm <- Simm %>% mutate(event=if_else((days_DeathTxCensor>=start & days_DeathTxCensor<=end), 1, 0))

Now will add date intervals for 5yr start and end times)

Simm <- Simm %>% mutate(end_5yr=(dx_date + days(end)))
Simm <- Simm %>% mutate(start_5yr=(end_5yr - days(1826)))

Now need to remove any rows where end_5yr is > date_DeathTxCensor (Amanda indicated need to remove rows where start_5yr > date_DeathTxCensor, but I don’t think this is correct because then we’d be assigning exposures to patients where they were alive/not transplanted for <5yrs for that exposure - need to double check with her)

Simm <- Simm %>% filter(!(end_5yr>DeathTxCensor_date & event!=1))

5 Simmons - Creating Time-Weighted Intervals of Exposures

PM <- PM %>% mutate(start=as.IDate(PM_date))
PM <- PM %>% mutate(end=as.IDate(PM_date + months(1) - days(1)))
PM <- as.data.table(PM)
str(PM)

## Classes 'data.table' and 'data.frame':   323248 obs. of  5 variables:
##  $ ID     : num  1097 1097 1097 1097 1097 ...
##  $ PM_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  12.7 17.1 9.6 8.2 14.7 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

Creating a list of intervals we want to calculate exposures for:

Simm_intervals <- Simm %>% dplyr::select(ID, start_5yr, end_5yr)
Simm_intervals <- Simm_intervals %>% mutate_at(c("start_5yr", "end_5yr"), as.IDate)
Simm_intervals <- Simm_intervals %>% rename(c("start"="start_5yr", "end"="end_5yr"))
Simm_intervals <- as.data.table(Simm_intervals)
str(Simm_intervals)

## Classes 'data.table' and 'data.frame':   89507 obs. of  3 variables:
##  $ ID   : num  1097 1097 1097 1097 1097 ...
##  $ start: IDate, format: "1997-02-23" "1997-03-25" ...
##  $ end  : IDate, format: "2002-02-23" "2002-03-25" ...
##  - attr(*, ".internal.selfref")=<externalptr>

PM_5yrWtedAvg <- intervalaverage(x=PM,
                y=Simm_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

PM_5yrWtedAvgx <- PM_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
PM_5yrWtedAvgx <- PM_5yrWtedAvgx %>% rename("PM"="value", "start_5yr"="start", "end_5yr"="end")
PM_5yrWtedAvgx <- PM_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(PM_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   89507 obs. of  4 variables:
##  $ ID       : num  1097 1097 1097 1097 1097 ...
##  $ PM       : num  12.5 12.4 12.3 12.3 12.5 ...
##  $ start_5yr: Date, format: "1997-02-23" "1997-03-25" ...
##  $ end_5yr  : Date, format: "2002-02-23" "2002-03-25" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to Simm

Simm <- left_join(Simm, PM_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

6 Simmons SO4 - Importing SO4 Datasets

Here I am importing the file which contains monthly SO4 level estimates by satellite at nearest lon/lat to Simmons patient residential addresses. These are linked to the patient ID.

outfile1 <- here("Simmons_fILD_2000_2017_SO4_2021_11_07.xlsx")
SO4 <- read_excel(outfile1)

7 Simmons SO4 - Simplifying SO4 Dataframe

SO4 <- SO4 %>% dplyr::select(!c(nrow, dist, lon, lat))

SO4 <- SO4 %>% 
  pivot_longer(cols=c(2:217), names_to="SO4_date", names_prefix="SO4_")

SO4x <- SO4 
SO4x$SO4_date <- gsub("jan", "01-01-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("feb", "01-02-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("mar", "01-03-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("apr", "01-04-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("may", "01-05-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("jun", "01-06-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("jul", "01-07-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("aug", "01-08-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("sep", "01-09-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("oct", "01-10-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("nov", "01-11-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("dec", "01-12-20", SO4x$SO4_date)

SO4x$SO4_date <- format(as.Date(SO4x$SO4_date, format="%d-%m-%Y"),"%Y-%m-%d")
SO4x$SO4_date <- as.Date(SO4x$SO4_date)
SO4 <- SO4x
rm(SO4x)

8 Simmons SO4 - Creating Time-Weighted Intervals of Exposures

SO4 <- SO4 %>% mutate(start=as.IDate(SO4_date))
SO4 <- SO4 %>% mutate(end=as.IDate(SO4_date + months(1) - days(1)))
SO4 <- as.data.table(SO4)
str(SO4)

## Classes 'data.table' and 'data.frame':   307584 obs. of  5 variables:
##  $ ID      : num  1097 1097 1097 1097 1097 ...
##  $ SO4_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value   : num  2.6 4.3 2.9 2.6 6.1 ...
##  $ start   : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end     : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

SO4_5yrWtedAvg <- intervalaverage(x=SO4,
                y=Simm_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

SO4_5yrWtedAvgx <- SO4_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
SO4_5yrWtedAvgx <- SO4_5yrWtedAvgx %>% rename("SO4"="value", "start_5yr"="start", "end_5yr"="end")
SO4_5yrWtedAvgx <- SO4_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(SO4_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   89507 obs. of  4 variables:
##  $ ID       : num  1097 1097 1097 1097 1097 ...
##  $ SO4      : num  4.41 4.37 4.35 4.33 4.41 ...
##  $ start_5yr: Date, format: "1997-02-23" "1997-03-25" ...
##  $ end_5yr  : Date, format: "2002-02-23" "2002-03-25" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to Simm

Simm <- left_join(Simm, SO4_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

9 Simmons NO3 - Importing NO3 Datasets

Here I am importing the file which contains monthly NO3 level estimates by satellite at nearest lon/lat to Simmons patient residential addresses. These are linked to the patient ID.

outfile1 <- here("Simm_fILD_2000_2017_NO3_2021_11_05.xlsx")
NO3 <- read_excel(outfile1)

10 Simmons NO3 - Simplifying NO3 Dataframe

NO3 <- NO3 %>% dplyr::select(!c(nrow, dist, lon, lat))

NO3 <- NO3 %>% 
  pivot_longer(cols=c(2:217), names_to="NO3_date", names_prefix="NIT_")

NO3x <- NO3 
NO3x$NO3_date <- gsub("jan", "01-01-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("feb", "01-02-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("mar", "01-03-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("apr", "01-04-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("may", "01-05-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("jun", "01-06-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("jul", "01-07-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("aug", "01-08-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("sep", "01-09-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("oct", "01-10-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("nov", "01-11-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("dec", "01-12-20", NO3x$NO3_date)

NO3x$NO3_date <- format(as.Date(NO3x$NO3_date, format="%d-%m-%Y"),"%Y-%m-%d")
NO3x$NO3_date <- as.Date(NO3x$NO3_date)
NO3 <- NO3x
rm(NO3x)

11 Simmons NO3 - Creating Time-Weighted Intervals of Exposures

NO3 <- NO3 %>% mutate(start=as.IDate(NO3_date))
NO3 <- NO3 %>% mutate(end=as.IDate(NO3_date + months(1) - days(1)))
NO3 <- as.data.table(NO3)
str(NO3)

## Classes 'data.table' and 'data.frame':   307584 obs. of  5 variables:
##  $ ID      : num  1097 1097 1097 1097 1097 ...
##  $ NO3_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value   : num  2.2 3.3 1.3 0.8 1.1 ...
##  $ start   : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end     : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

NO3_5yrWtedAvg <- intervalaverage(x=NO3,
                y=Simm_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

NO3_5yrWtedAvgx <- NO3_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
NO3_5yrWtedAvgx <- NO3_5yrWtedAvgx %>% rename("NO3"="value", "start_5yr"="start", "end_5yr"="end")
NO3_5yrWtedAvgx <- NO3_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(NO3_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   89507 obs. of  4 variables:
##  $ ID       : num  1097 1097 1097 1097 1097 ...
##  $ NO3      : num  1.4 1.41 1.4 1.38 1.36 ...
##  $ start_5yr: Date, format: "1997-02-23" "1997-03-25" ...
##  $ end_5yr  : Date, format: "2002-02-23" "2002-03-25" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to Simm

Simm <- left_join(Simm, NO3_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

12 Simmons NH4 - Importing NH4 Datasets

Here I am importing the file which contains monthly NH4 level estimates by satellite at nearest lon/lat to Simmons patient residential addresses. These are linked to the patient ID.

outfile1 <- here("Simm_fILD_2000_2017_NH4_2021_11_05.xlsx")
NH4 <- read_excel(outfile1)

13 Simmons NH4 - Simplifying NH4 Dataframe

NH4 <- NH4 %>% dplyr::select(!c(nrow, dist, lon, lat))

NH4 <- NH4 %>% 
  pivot_longer(cols=c(2:217), names_to="NH4_date", names_prefix="NH4_")

NH4x <- NH4 
NH4x$NH4_date <- gsub("jan", "01-01-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("feb", "01-02-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("mar", "01-03-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("apr", "01-04-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("may", "01-05-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("jun", "01-06-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("jul", "01-07-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("aug", "01-08-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("sep", "01-09-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("oct", "01-10-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("nov", "01-11-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("dec", "01-12-20", NH4x$NH4_date)

NH4x$NH4_date <- format(as.Date(NH4x$NH4_date, format="%d-%m-%Y"),"%Y-%m-%d")
NH4x$NH4_date <- as.Date(NH4x$NH4_date)
NH4 <- NH4x
rm(NH4x)

14 Simmons NH4 - Creating Time-Weighted Intervals of Exposures

NH4 <- NH4 %>% mutate(start=as.IDate(NH4_date))
NH4 <- NH4 %>% mutate(end=as.IDate(NH4_date + months(1) - days(1)))
NH4 <- as.data.table(NH4)
str(NH4)

## Classes 'data.table' and 'data.frame':   307584 obs. of  5 variables:
##  $ ID      : num  1097 1097 1097 1097 1097 ...
##  $ NH4_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value   : num  1.5 2.7 1.1 1.2 2.5 ...
##  $ start   : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end     : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

NH4_5yrWtedAvg <- intervalaverage(x=NH4,
                y=Simm_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

NH4_5yrWtedAvgx <- NH4_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
NH4_5yrWtedAvgx <- NH4_5yrWtedAvgx %>% rename("NH4"="value", "start_5yr"="start", "end_5yr"="end")
NH4_5yrWtedAvgx <- NH4_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(NH4_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   89507 obs. of  4 variables:
##  $ ID       : num  1097 1097 1097 1097 1097 ...
##  $ NH4      : num  1.7 1.69 1.68 1.67 1.69 ...
##  $ start_5yr: Date, format: "1997-02-23" "1997-03-25" ...
##  $ end_5yr  : Date, format: "2002-02-23" "2002-03-25" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to Simm

Simm <- left_join(Simm, NH4_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

15 Simmons BC - Importing BC Datasets

Here I am importing the file which contains monthly BC level estimates by satellite at nearest lon/lat to Simmons patient residential addresses. These are linked to the patient ID.

outfile1 <- here("Simm_fILD_2000_2017_BC_2021_11_05.xlsx")
BC <- read_excel(outfile1)

16 Simmons BC - Simplifying BC Dataframe

BC <- BC %>% dplyr::select(!c(nrow, dist, lon, lat))

BC <- BC %>% 
  pivot_longer(cols=c(2:217), names_to="BC_date", names_prefix="BC_")

BCx <- BC 
BCx$BC_date <- gsub("jan", "01-01-20", BCx$BC_date)
BCx$BC_date <- gsub("feb", "01-02-20", BCx$BC_date)
BCx$BC_date <- gsub("mar", "01-03-20", BCx$BC_date)
BCx$BC_date <- gsub("apr", "01-04-20", BCx$BC_date)
BCx$BC_date <- gsub("may", "01-05-20", BCx$BC_date)
BCx$BC_date <- gsub("jun", "01-06-20", BCx$BC_date)
BCx$BC_date <- gsub("jul", "01-07-20", BCx$BC_date)
BCx$BC_date <- gsub("aug", "01-08-20", BCx$BC_date)
BCx$BC_date <- gsub("sep", "01-09-20", BCx$BC_date)
BCx$BC_date <- gsub("oct", "01-10-20", BCx$BC_date)
BCx$BC_date <- gsub("nov", "01-11-20", BCx$BC_date)
BCx$BC_date <- gsub("dec", "01-12-20", BCx$BC_date)

BCx$BC_date <- format(as.Date(BCx$BC_date, format="%d-%m-%Y"),"%Y-%m-%d")
BCx$BC_date <- as.Date(BCx$BC_date)
BC <- BCx
rm(BCx)

17 Simmons BC - Creating Time-Weighted Intervals of Exposures

BC <- BC %>% mutate(start=as.IDate(BC_date))
BC <- BC %>% mutate(end=as.IDate(BC_date + months(1) - days(1)))
BC <- as.data.table(BC)
str(BC)

## Classes 'data.table' and 'data.frame':   307584 obs. of  5 variables:
##  $ ID     : num  1097 1097 1097 1097 1097 ...
##  $ BC_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  0.7 1 0.8 0.7 1.4 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

BC_5yrWtedAvg <- intervalaverage(x=BC,
                y=Simm_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

BC_5yrWtedAvgx <- BC_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
BC_5yrWtedAvgx <- BC_5yrWtedAvgx %>% rename("BC"="value", "start_5yr"="start", "end_5yr"="end")
BC_5yrWtedAvgx <- BC_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(BC_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   89507 obs. of  4 variables:
##  $ ID       : num  1097 1097 1097 1097 1097 ...
##  $ BC       : num  0.774 0.768 0.762 0.753 0.763 ...
##  $ start_5yr: Date, format: "1997-02-23" "1997-03-25" ...
##  $ end_5yr  : Date, format: "2002-02-23" "2002-03-25" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to Simm

Simm <- left_join(Simm, BC_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

18 Simmons OM - Importing OM Datasets

Here I am importing the file which contains monthly OM level estimates by satellite at nearest lon/lat to Simmons patient residential addresses. These are linked to the patient ID.

outfile1 <- here("Simm_fILD_2000_2017_OM_2021_11_05.xlsx")
OM <- read_excel(outfile1)

19 Simmons OM - Simplifying OM Dataframe

OM <- OM %>% dplyr::select(!c(nrow, dist, lon, lat))

OM <- OM %>% 
  pivot_longer(cols=c(2:217), names_to="OM_date", names_prefix="OM_")

OMx <- OM 
OMx$OM_date <- gsub("jan", "01-01-20", OMx$OM_date)
OMx$OM_date <- gsub("feb", "01-02-20", OMx$OM_date)
OMx$OM_date <- gsub("mar", "01-03-20", OMx$OM_date)
OMx$OM_date <- gsub("apr", "01-04-20", OMx$OM_date)
OMx$OM_date <- gsub("may", "01-05-20", OMx$OM_date)
OMx$OM_date <- gsub("jun", "01-06-20", OMx$OM_date)
OMx$OM_date <- gsub("jul", "01-07-20", OMx$OM_date)
OMx$OM_date <- gsub("aug", "01-08-20", OMx$OM_date)
OMx$OM_date <- gsub("sep", "01-09-20", OMx$OM_date)
OMx$OM_date <- gsub("oct", "01-10-20", OMx$OM_date)
OMx$OM_date <- gsub("nov", "01-11-20", OMx$OM_date)
OMx$OM_date <- gsub("dec", "01-12-20", OMx$OM_date)

OMx$OM_date <- format(as.Date(OMx$OM_date, format="%d-%m-%Y"),"%Y-%m-%d")
OMx$OM_date <- as.Date(OMx$OM_date)
OM <- OMx
rm(OMx)

20 Simmons OM - Creating Time-Weighted Intervals of Exposures

OM <- OM %>% mutate(start=as.IDate(OM_date))
OM <- OM %>% mutate(end=as.IDate(OM_date + months(1) - days(1)))
OM <- as.data.table(OM)
str(OM)

## Classes 'data.table' and 'data.frame':   307584 obs. of  5 variables:
##  $ ID     : num  1097 1097 1097 1097 1097 ...
##  $ OM_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  3 3.2 3.3 2 3.7 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

OM_5yrWtedAvg <- intervalaverage(x=OM,
                y=Simm_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

OM_5yrWtedAvgx <- OM_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
OM_5yrWtedAvgx <- OM_5yrWtedAvgx %>% rename("OM"="value", "start_5yr"="start", "end_5yr"="end")
OM_5yrWtedAvgx <- OM_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(OM_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   89507 obs. of  4 variables:
##  $ ID       : num  1097 1097 1097 1097 1097 ...
##  $ OM       : num  2.9 2.89 2.88 2.85 2.9 ...
##  $ start_5yr: Date, format: "1997-02-23" "1997-03-25" ...
##  $ end_5yr  : Date, format: "2002-02-23" "2002-03-25" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to Simm

Simm <- left_join(Simm, OM_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

21 Simmons SS - Importing SS Datasets

Here I am importing the file which contains monthly SS level estimates by satellite at nearest lon/lat to Simmons patient residential addresses. These are linked to the patient ID.

outfile1 <- here("Simm_fILD_2000_2017_SS_2021_11_05.xlsx")
SS <- read_excel(outfile1)

22 Simmons SS - Simplifying SS Dataframe

SS <- SS %>% dplyr::select(!c(nrow, dist, lon, lat))

SS <- SS %>% 
  pivot_longer(cols=c(2:217), names_to="SS_date", names_prefix="SS_")

SSx <- SS 
SSx$SS_date <- gsub("jan", "01-01-20", SSx$SS_date)
SSx$SS_date <- gsub("feb", "01-02-20", SSx$SS_date)
SSx$SS_date <- gsub("mar", "01-03-20", SSx$SS_date)
SSx$SS_date <- gsub("apr", "01-04-20", SSx$SS_date)
SSx$SS_date <- gsub("may", "01-05-20", SSx$SS_date)
SSx$SS_date <- gsub("jun", "01-06-20", SSx$SS_date)
SSx$SS_date <- gsub("jul", "01-07-20", SSx$SS_date)
SSx$SS_date <- gsub("aug", "01-08-20", SSx$SS_date)
SSx$SS_date <- gsub("sep", "01-09-20", SSx$SS_date)
SSx$SS_date <- gsub("oct", "01-10-20", SSx$SS_date)
SSx$SS_date <- gsub("nov", "01-11-20", SSx$SS_date)
SSx$SS_date <- gsub("dec", "01-12-20", SSx$SS_date)

SSx$SS_date <- format(as.Date(SSx$SS_date, format="%d-%m-%Y"),"%Y-%m-%d")
SSx$SS_date <- as.Date(SSx$SS_date)
SS <- SSx
rm(SSx)

23 Simmons SS - Creating Time-Weighted Intervals of Exposures

SS <- SS %>% mutate(start=as.IDate(SS_date))
SS <- SS %>% mutate(end=as.IDate(SS_date + months(1) - days(1)))
SS <- as.data.table(SS)
str(SS)

## Classes 'data.table' and 'data.frame':   307584 obs. of  5 variables:
##  $ ID     : num  1097 1097 1097 1097 1097 ...
##  $ SS_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  0 0.8 0.5 0 0 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

SS_5yrWtedAvg <- intervalaverage(x=SS,
                y=Simm_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

SS_5yrWtedAvgx <- SS_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
SS_5yrWtedAvgx <- SS_5yrWtedAvgx %>% rename("SS"="value", "start_5yr"="start", "end_5yr"="end")
SS_5yrWtedAvgx <- SS_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(SS_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   89507 obs. of  4 variables:
##  $ ID       : num  1097 1097 1097 1097 1097 ...
##  $ SS       : num  0.241 0.248 0.242 0.248 0.249 ...
##  $ start_5yr: Date, format: "1997-02-23" "1997-03-25" ...
##  $ end_5yr  : Date, format: "2002-02-23" "2002-03-25" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to Simm

Simm <- left_join(Simm, SS_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

24 Simmons Soil - Importing Soil Datasets

Here I am importing the file which contains monthly Soil level estimates by satellite at nearest lon/lat to Simmons patient residential addresses. These are linked to the patient ID.

outfile1 <- here("Simm_fILD_2000_2017_Soil_2021_11_05.xlsx")
Soil <- read_excel(outfile1)

25 Simmons Soil - Simplifying Soil Dataframe

Soil <- Soil %>% dplyr::select(!c(nrow, dist, lon, lat))

Soil <- Soil %>% 
  pivot_longer(cols=c(2:217), names_to="Soil_date", names_prefix="soil_")

Soilx <- Soil 
Soilx$Soil_date <- gsub("jan", "01-01-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("feb", "01-02-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("mar", "01-03-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("apr", "01-04-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("may", "01-05-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("jun", "01-06-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("jul", "01-07-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("aug", "01-08-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("sep", "01-09-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("oct", "01-10-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("nov", "01-11-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("dec", "01-12-20", Soilx$Soil_date)

Soilx$Soil_date <- format(as.Date(Soilx$Soil_date, format="%d-%m-%Y"),"%Y-%m-%d")
Soilx$Soil_date <- as.Date(Soilx$Soil_date)
Soil <- Soilx
rm(Soilx)

26 Simmons Soil - Creating Time-Weighted Intervals of Exposures

Soil <- Soil %>% mutate(start=as.IDate(Soil_date))
Soil <- Soil %>% mutate(end=as.IDate(Soil_date + months(1) - days(1)))
Soil <- as.data.table(Soil)
str(Soil)

## Classes 'data.table' and 'data.frame':   307584 obs. of  5 variables:
##  $ ID       : num  1097 1097 1097 1097 1097 ...
##  $ Soil_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value    : num  0.4 0.8 0.7 0.3 0.7 ...
##  $ start    : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end      : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

Soil_5yrWtedAvg <- intervalaverage(x=Soil,
                y=Simm_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

Soil_5yrWtedAvgx <- Soil_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
Soil_5yrWtedAvgx <- Soil_5yrWtedAvgx %>% rename("Soil"="value", "start_5yr"="start", "end_5yr"="end")
Soil_5yrWtedAvgx <- Soil_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(Soil_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   89507 obs. of  4 variables:
##  $ ID       : num  1097 1097 1097 1097 1097 ...
##  $ Soil     : num  0.521 0.513 0.511 0.511 0.515 ...
##  $ start_5yr: Date, format: "1997-02-23" "1997-03-25" ...
##  $ end_5yr  : Date, format: "2002-02-23" "2002-03-25" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to Simm

Simm <- left_join(Simm, Soil_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

27 PFF - Importing Datasets

Here I am importing the file which contains monthly PM level estimates by satellite at nearest lon/lat to PFF patient residential addresses. These are linked to the patient ID.

outfile1 <- here("PFF_fILD_2000_2018_PM25_2021_10_08.xlsx")
PM <- read_excel(outfile1)
PM <- PM %>% rename("SSID"="ID")

Match up PM ID to SSID from matching file

outfile2 <- here("PFF_fILD_PM25_BaselineData_2021_10_20.xlsx")
PM25 <- read_excel(outfile2)
PM25 <- PM25 %>% dplyr::select(ID, SSID)

Merge PM and PM

PM <- left_join(PM, PM25, by="SSID")

Here I am importing the file containing the complete baseline clinical and demographic data for 1905 PFF patients

outfile3 <- here("PFF_fILDPts_BaselineData_ConsentDateReference_2022_08_01.xlsx")
PFF <- read_excel(outfile3)

28 PFF - Merging Datasets

I used a inner_join here so that the complete “PFF” dataframe only includes patients with fILD that have all baseline demographics and PM data.

PFF <- inner_join(PFF, PM, by="SSID")

This results in 1905 complete records

Reorder so “ID” is the first column

PFF <- PFF %>% dplyr::select(ID, everything(.))

29 PFF - Making dx_yr column

PFF <- PFF %>% 
  mutate(dx_yr = format(as.Date(PFF$dx_date, format="%Y-%m-%d"),"%Y"))
PFF$dx_yr <- as.numeric(PFF$dx_yr)

30 PFF - Creating Site Variable

PFF$site <- substr(PFF$SSID, 1,3)
PFF$site <- as.factor(PFF$site)
str(PFF$site)

##  Factor w/ 42 levels "01R","02R","03R",..: 24 2 3 3 3 3 3 4 4 4 ...

Removing Pittsburgh site

PFF <- PFF %>% filter(!site=="08R")

31 PFF - Simplifying PM Dataframe

Reorder so “ID” is the first column

PM <- PM %>% dplyr::select(ID, everything(.))
PM <- PM %>% dplyr::select(!SSID)

PM <- PM %>% dplyr::select(!c(nrow, dist, lon, lat))
colnames(PM)

##   [1] "ID"         "PM25_jan00" "PM25_feb00" "PM25_mar00" "PM25_apr00"
##   [6] "PM25_may00" "PM25_jun00" "PM25_jul00" "PM25_aug00" "PM25_sep00"
##  [11] "PM25_oct00" "PM25_nov00" "PM25_dec00" "PM25_jan01" "PM25_feb01"
##  [16] "PM25_mar01" "PM25_apr01" "PM25_may01" "PM25_jun01" "PM25_jul01"
##  [21] "PM25_aug01" "PM25_sep01" "PM25_oct01" "PM25_nov01" "PM25_dec01"
##  [26] "PM25_jan02" "PM25_feb02" "PM25_mar02" "PM25_apr02" "PM25_may02"
##  [31] "PM25_jun02" "PM25_jul02" "PM25_aug02" "PM25_sep02" "PM25_oct02"
##  [36] "PM25_nov02" "PM25_dec02" "PM25_jan03" "PM25_feb03" "PM25_mar03"
##  [41] "PM25_apr03" "PM25_may03" "PM25_jun03" "PM25_jul03" "PM25_aug03"
##  [46] "PM25_sep03" "PM25_oct03" "PM25_nov03" "PM25_dec03" "PM25_jan04"
##  [51] "PM25_feb04" "PM25_mar04" "PM25_apr04" "PM25_may04" "PM25_jun04"
##  [56] "PM25_jul04" "PM25_aug04" "PM25_sep04" "PM25_oct04" "PM25_nov04"
##  [61] "PM25_dec04" "PM25_jan05" "PM25_feb05" "PM25_mar05" "PM25_apr05"
##  [66] "PM25_may05" "PM25_jun05" "PM25_jul05" "PM25_aug05" "PM25_sep05"
##  [71] "PM25_oct05" "PM25_nov05" "PM25_dec05" "PM25_jan06" "PM25_feb06"
##  [76] "PM25_mar06" "PM25_apr06" "PM25_may06" "PM25_jun06" "PM25_jul06"
##  [81] "PM25_aug06" "PM25_sep06" "PM25_oct06" "PM25_nov06" "PM25_dec06"
##  [86] "PM25_jan07" "PM25_feb07" "PM25_mar07" "PM25_apr07" "PM25_may07"
##  [91] "PM25_jun07" "PM25_jul07" "PM25_aug07" "PM25_sep07" "PM25_oct07"
##  [96] "PM25_nov07" "PM25_dec07" "PM25_jan08" "PM25_feb08" "PM25_mar08"
## [101] "PM25_apr08" "PM25_may08" "PM25_jun08" "PM25_jul08" "PM25_aug08"
## [106] "PM25_sep08" "PM25_oct08" "PM25_nov08" "PM25_dec08" "PM25_jan09"
## [111] "PM25_feb09" "PM25_mar09" "PM25_apr09" "PM25_may09" "PM25_jun09"
## [116] "PM25_jul09" "PM25_aug09" "PM25_sep09" "PM25_oct09" "PM25_nov09"
## [121] "PM25_dec09" "PM25_jan10" "PM25_feb10" "PM25_mar10" "PM25_apr10"
## [126] "PM25_may10" "PM25_jun10" "PM25_jul10" "PM25_aug10" "PM25_sep10"
## [131] "PM25_oct10" "PM25_nov10" "PM25_dec10" "PM25_jan11" "PM25_feb11"
## [136] "PM25_mar11" "PM25_apr11" "PM25_may11" "PM25_jun11" "PM25_jul11"
## [141] "PM25_aug11" "PM25_sep11" "PM25_oct11" "PM25_nov11" "PM25_dec11"
## [146] "PM25_jan12" "PM25_feb12" "PM25_mar12" "PM25_apr12" "PM25_may12"
## [151] "PM25_jun12" "PM25_jul12" "PM25_aug12" "PM25_sep12" "PM25_oct12"
## [156] "PM25_nov12" "PM25_dec12" "PM25_jan13" "PM25_feb13" "PM25_mar13"
## [161] "PM25_apr13" "PM25_may13" "PM25_jun13" "PM25_jul13" "PM25_aug13"
## [166] "PM25_sep13" "PM25_oct13" "PM25_nov13" "PM25_dec13" "PM25_jan14"
## [171] "PM25_feb14" "PM25_mar14" "PM25_apr14" "PM25_may14" "PM25_jun14"
## [176] "PM25_jul14" "PM25_aug14" "PM25_sep14" "PM25_oct14" "PM25_nov14"
## [181] "PM25_dec14" "PM25_jan15" "PM25_feb15" "PM25_mar15" "PM25_apr15"
## [186] "PM25_may15" "PM25_jun15" "PM25_jul15" "PM25_aug15" "PM25_sep15"
## [191] "PM25_oct15" "PM25_nov15" "PM25_dec15" "PM25_jan16" "PM25_feb16"
## [196] "PM25_mar16" "PM25_apr16" "PM25_may16" "PM25_jun16" "PM25_jul16"
## [201] "PM25_aug16" "PM25_sep16" "PM25_oct16" "PM25_nov16" "PM25_dec16"
## [206] "PM25_jan17" "PM25_feb17" "PM25_mar17" "PM25_apr17" "PM25_may17"
## [211] "PM25_jun17" "PM25_jul17" "PM25_aug17" "PM25_sep17" "PM25_oct17"
## [216] "PM25_nov17" "PM25_dec17" "PM25_jan18" "PM25_feb18" "PM25_mar18"
## [221] "PM25_apr18" "PM25_may18" "PM25_jun18" "PM25_jul18" "PM25_aug18"
## [226] "PM25_sep18" "PM25_oct18" "PM25_nov18" "PM25_dec18"

PM <- PM %>% 
  pivot_longer(cols=c(2:228), names_to="PM_date", names_prefix="PM25_")

PMx <- PM 
PMx$PM_date <- gsub("jan", "01-01-20", PMx$PM_date)
PMx$PM_date <- gsub("feb", "01-02-20", PMx$PM_date)
PMx$PM_date <- gsub("mar", "01-03-20", PMx$PM_date)
PMx$PM_date <- gsub("apr", "01-04-20", PMx$PM_date)
PMx$PM_date <- gsub("may", "01-05-20", PMx$PM_date)
PMx$PM_date <- gsub("jun", "01-06-20", PMx$PM_date)
PMx$PM_date <- gsub("jul", "01-07-20", PMx$PM_date)
PMx$PM_date <- gsub("aug", "01-08-20", PMx$PM_date)
PMx$PM_date <- gsub("sep", "01-09-20", PMx$PM_date)
PMx$PM_date <- gsub("oct", "01-10-20", PMx$PM_date)
PMx$PM_date <- gsub("nov", "01-11-20", PMx$PM_date)
PMx$PM_date <- gsub("dec", "01-12-20", PMx$PM_date)

PMx$PM_date <- format(as.Date(PMx$PM_date, format="%d-%m-%Y"),"%Y-%m-%d")
PMx$PM_date <- as.Date(PMx$PM_date)
PM <- PMx
rm(PMx)

32 PFF - Simplifying PFF Dataframe

32.1 PFF - Death/Transplant/Censor Date

Extracting year of diagnosis and year of death/transplant/censoring

#Start with the year of diagnosis
PFF <- PFF %>% 
  mutate(dx_yrmo = format(as.Date(PFF$dx_date, format="%Y-%m-%d"),"%Y-%m"))
PFF <- PFF %>% 
  mutate(dx_yr = format(as.Date(PFF$dx_date, format="%Y-%m-%d"),"%Y"))
PFF$dx_yr <- as.numeric(PFF$dx_yr)

#Then the year of death or lung transplant
PFF <- PFF %>% 
  mutate(deathORtx_date = if_else(!is.na(tx_date), tx_date, death_date))
PFF <- PFF %>% 
  mutate(deathORtx_yrmo = format(as.Date(PFF$deathORtx_date, format="%Y-%m-%d"),"%Y-%m"))

#Then the year the records were last updated (i.e. year of censoring)
PFF <- PFF %>% 
  mutate(DeathTxCensor_date = if_else(!is.na(deathORtx_date), deathORtx_date, censor_date))
PFF <- PFF %>% 
  mutate(censor_yrmo = format(as.Date(PFF$DeathTxCensor_date, format="%Y-%m-%d"),"%Y-%m"))

32.2 PFF - Removing Unnecessary Columns and Correcting Dates

PFF <- PFF %>% dplyr::select(c(ID, site, age_dx, sex, smokeHx, race, dich_Race, pct_belowpoverty, dx_group, dx, dx_date, death_date, tx_date, DeathTxCensor_date, censor_date, fvc_date, dlco_date, fvc_pct, dlco_pct, status, deadORtx, time_DeathTxCensor, dx_yr))
PFF <- PFF %>% 
  mutate_at(c("death_date", "censor_date", "tx_date", "dx_date", "fvc_date", "dlco_date", "DeathTxCensor_date"), as.Date)
str(PFF)

## tibble [1,870 × 23] (S3: tbl_df/tbl/data.frame)
##  $ ID                : num [1:1870] 1 2 3 4 5 6 7 8 9 10 ...
##  $ site              : Factor w/ 42 levels "01R","02R","03R",..: 24 2 3 3 3 3 3 4 4 4 ...
##  $ age_dx            : num [1:1870] 64.5 42.7 49.6 70.6 50.6 ...
##  $ sex               : chr [1:1870] "Female" "Female" "Female" "Female" ...
##  $ smokeHx           : chr [1:1870] "Never" "Never" "Never" "Never" ...
##  $ race              : chr [1:1870] "W" "W" "W" "W" ...
##  $ dich_Race         : chr [1:1870] "White" "White" "White" "White" ...
##  $ pct_belowpoverty  : num [1:1870] 3.4 7.7 9.3 8.4 9.1 13.7 4.9 6.5 5.9 10.6 ...
##  $ dx_group          : chr [1:1870] "CTD-ILD" "CTD-ILD" "CTD-ILD" "CTD-ILD" ...
##  $ dx                : chr [1:1870] "ANKSPON_ILD" "ANTISYNTHETASE" "ANTISYNTHETASE" "ANTISYNTHETASE" ...
##  $ dx_date           : Date[1:1870], format: "2006-07-07" "2017-06-16" ...
##  $ death_date        : Date[1:1870], format: NA NA ...
##  $ tx_date           : Date[1:1870], format: NA NA ...
##  $ DeathTxCensor_date: Date[1:1870], format: "2021-03-03" "2019-04-01" ...
##  $ censor_date       : Date[1:1870], format: "2021-03-03" "2019-04-01" ...
##  $ fvc_date          : Date[1:1870], format: "2016-09-23" "2017-08-25" ...
##  $ dlco_date         : Date[1:1870], format: "2016-09-23" "2017-08-25" ...
##  $ fvc_pct           : num [1:1870] 38.1 72.9 85 79.4 86.1 ...
##  $ dlco_pct          : num [1:1870] 80.4 80.8 105.2 41.2 82.3 ...
##  $ status            : chr [1:1870] "0" "0" "0" "0" ...
##  $ deadORtx          : num [1:1870] 0 0 0 0 0 0 0 0 0 0 ...
##  $ time_DeathTxCensor: num [1:1870] 4.44 1.8 3.95 2.64 2.64 ...
##  $ dx_yr             : num [1:1870] 2006 2017 2016 2017 2017 ...

32.3 PFF - Correcting Smoking

Need to correct smoking variables

PFF$smokeHx <- as.character(PFF$smokeHx) 
PFF <- PFF %>% mutate(smokeHx1=if_else(is.na(smokeHx), "Unknown", smokeHx))

Now need to remove old smoking variables and rename new ones

PFF <- PFF %>% dplyr::select(-c(smokeHx))
PFF <- PFF %>% rename(c("smokeHx"="smokeHx1"))
PFF$smokeHx <- as.factor(PFF$smokeHx)
PFF$smokeHx <- fct_relevel(PFF$smokeHx, c("Never","Ever","Unknown"))

## Warning: 1 unknown level in `f`: Unknown

32.4 PFF - Correcting Factors

Need to correct other factor variables

PFF$sex <- fct_relevel(PFF$sex, c("M","F"))

## Warning: 2 unknown levels in `f`: M and F

PFF$race <- fct_relevel(PFF$race, c("W","B","A","N","U"))

## Warning: 1 unknown level in `f`: N

PFF$dich_Race <- fct_relevel(PFF$dich_Race, c("White","Non-White"))
PFF$dx <- fct_relevel(PFF$dx, c("IPF"))
PFF$dx_group <- fct_relevel(PFF$dx_group, c("IPF"))
PFF <- PFF %>% mutate_at(c("status", "site"), as.factor)

32.5 PFF - Creating New Variables

32.5.1 PFF - IPF vs Other Diagnosis

PFF <- PFF %>% mutate(dx_IPF=ifelse(dx=="IPF", "IPF", "not_IPF"))  
PFF$dx_IPF <- fct_relevel(PFF$dx_IPF, c("IPF"))

32.5.2 PFF - Disadvantage Distribution

Creating this empirical cumulative distribution will allow us to combine the analyses of all three cohorts even though the measurements for disadvantage are different between the three.

plot(ecdf(PFF$pct_belowpoverty))

PFF$disadv <- ecdf(PFF$pct_belowpoverty)(PFF$pct_belowpoverty)

33 PFF - Modifying PFF Dataset

Take down PFF dataset to the 1905 patients with complete data

IDs <- as.data.table(unique(PM$ID))
IDs <- IDs %>% rename("ID"="V1")
PFF <- left_join(IDs, PFF, by="ID")

PFF <- PFF %>% mutate(days_DeathTxCensor=(time_DeathTxCensor*365.25))

Longest time_DeathTxCensor= 1848days

30-day longest time interval would be 1860 days

start <- seq(1, 1831, by = 30)
start

##  [1]    1   31   61   91  121  151  181  211  241  271  301  331  361  391  421
## [16]  451  481  511  541  571  601  631  661  691  721  751  781  811  841  871
## [31]  901  931  961  991 1021 1051 1081 1111 1141 1171 1201 1231 1261 1291 1321
## [46] 1351 1381 1411 1441 1471 1501 1531 1561 1591 1621 1651 1681 1711 1741 1771
## [61] 1801 1831

end <- seq(30, 1860, by = 30)
end

##  [1]   30   60   90  120  150  180  210  240  270  300  330  360  390  420  450
## [16]  480  510  540  570  600  630  660  690  720  750  780  810  840  870  900
## [31]  930  960  990 1020 1050 1080 1110 1140 1170 1200 1230 1260 1290 1320 1350
## [46] 1380 1410 1440 1470 1500 1530 1560 1590 1620 1650 1680 1710 1740 1770 1800
## [61] 1830 1860

Repeat the list of intervals 1905 times (number of patients in PFF)

start <- rep(start, times=1905)
end <- rep(end, times=1905)
intervals <- as.data.frame(cbind(start, end))

Add ID column to intervals

IDs <- rep(PFF$ID, each=62)
intervals <- as.data.frame(cbind(IDs, intervals))
intervals <- intervals %>% rename("ID"="IDs")

Join PFF and intervals

PFF <- left_join(intervals, PFF, by="ID")

Determine if event occurred during interval

PFF <- PFF %>% mutate(event=if_else((days_DeathTxCensor>=start & days_DeathTxCensor<=end), 1, 0))

Now will add date intervals for 5yr start and end times)

PFF <- PFF %>% mutate(end_5yr=(dx_date + days(end)))
PFF <- PFF %>% mutate(start_5yr=(end_5yr - days(1826)))

Now need to remove any rows where end_5yr is > date_DeathTxCensor (Amanda indicated need to remove rows where start_5yr > date_DeathTxCensor, but I don’t think this is correct because then we’d be assigning exposures to patients where they were alive/not transplanted for <5yrs for that exposure - need to double check with her)

PFF <- PFF %>% filter(!(end_5yr>DeathTxCensor_date & event!=1))

34 PFF - Creating Time-Weighted Intervals of Exposures

PM <- PM %>% mutate(start=as.IDate(PM_date))
PM <- PM %>% mutate(end=as.IDate(PM_date + months(1) - days(1)))
PM <- as.data.table(PM)
str(PM)

## Classes 'data.table' and 'data.frame':   432435 obs. of  6 variables:
##  $ ID        : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ PM25_dec18: num  11.1 11.1 11.1 11.1 11.1 ...
##  $ PM_date   : Date, format: "2000-01-01" "2000-02-01" ...
##  $ value     : num  14.3 12.7 11.9 7.6 7.4 ...
##  $ start     : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end       : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

Creating a list of intervals we want to calculate exposures for:

PFF_intervals <- PFF %>% dplyr::select(ID, start_5yr, end_5yr)
PFF_intervals <- PFF_intervals %>% mutate_at(c("start_5yr", "end_5yr"), as.IDate)
PFF_intervals <- PFF_intervals %>% rename(c("start"="start_5yr", "end"="end_5yr"))
PFF_intervals <- as.data.table(PFF_intervals)
str(PFF_intervals)

## Classes 'data.table' and 'data.frame':   87626 obs. of  3 variables:
##  $ ID   : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ start: IDate, format: "2001-08-06" "2001-09-05" ...
##  $ end  : IDate, format: "2006-08-06" "2006-09-05" ...
##  - attr(*, ".internal.selfref")=<externalptr>

PM_5yrWtedAvg <- intervalaverage(x=PM,
                y=PFF_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

PM_5yrWtedAvgx <- PM_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
PM_5yrWtedAvgx <- PM_5yrWtedAvgx %>% rename("PM"="value", "start_5yr"="start", "end_5yr"="end")
PM_5yrWtedAvgx <- PM_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(PM_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   87626 obs. of  4 variables:
##  $ ID       : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ PM       : num  9.9 9.89 9.88 9.89 9.85 ...
##  $ start_5yr: Date, format: "2001-08-06" "2001-09-05" ...
##  $ end_5yr  : Date, format: "2006-08-06" "2006-09-05" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to PFF

PFF <- left_join(PFF, PM_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

35 PFF SO4 - Importing SO4 Datasets

Here I am importing the file which contains monthly SO4 level estimates by satellite at nearest lon/lat to PFF patient residential addresses. These are linked to the patient ID.

outfile1 <- here("PFF_fILD_2000_2017_SO4_2021_11_05.xlsx")
SO4 <- read_excel(outfile1)

36 PFF SO4 - Simplifying SO4 Dataframe

SO4 <- SO4 %>% dplyr::select(!c(nrow, dist, lon, lat))

SO4 <- SO4 %>% 
  pivot_longer(cols=c(2:217), names_to="SO4_date", names_prefix="SO4_")

SO4x <- SO4 
SO4x$SO4_date <- gsub("jan", "01-01-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("feb", "01-02-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("mar", "01-03-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("apr", "01-04-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("may", "01-05-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("jun", "01-06-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("jul", "01-07-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("aug", "01-08-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("sep", "01-09-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("oct", "01-10-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("nov", "01-11-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("dec", "01-12-20", SO4x$SO4_date)

SO4x$SO4_date <- format(as.Date(SO4x$SO4_date, format="%d-%m-%Y"),"%Y-%m-%d")
SO4x$SO4_date <- as.Date(SO4x$SO4_date)
SO4 <- SO4x
rm(SO4x)

37 PFF SO4 - Creating Time-Weighted Intervals of Exposures

SO4 <- SO4 %>% mutate(start=as.IDate(SO4_date))
SO4 <- SO4 %>% mutate(end=as.IDate(SO4_date + months(1) - days(1)))
SO4 <- as.data.table(SO4)
str(SO4)

## Classes 'data.table' and 'data.frame':   411480 obs. of  5 variables:
##  $ ID      : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ SO4_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value   : num  2.6 2.1 2.4 1.6 1.9 ...
##  $ start   : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end     : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

SO4_5yrWtedAvg <- intervalaverage(x=SO4,
                y=PFF_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

SO4_5yrWtedAvgx <- SO4_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
SO4_5yrWtedAvgx <- SO4_5yrWtedAvgx %>% rename("SO4"="value", "start_5yr"="start", "end_5yr"="end")
SO4_5yrWtedAvgx <- SO4_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(SO4_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   87626 obs. of  4 variables:
##  $ ID       : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ SO4      : num  2.3 2.3 2.28 2.29 2.27 ...
##  $ start_5yr: Date, format: "2001-08-06" "2001-09-05" ...
##  $ end_5yr  : Date, format: "2006-08-06" "2006-09-05" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to PFF

PFF <- left_join(PFF, SO4_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

38 PFF NO3 - Importing NO3 Datasets

Here I am importing the file which contains monthly NO3 level estimates by satellite at nearest lon/lat to PFF patient residential addresses. These are linked to the patient ID.

outfile1 <- here("PFF_fILD_2000_2017_NO3_2021_11_05.xlsx")
NO3 <- read_excel(outfile1)

39 PFF NO3 - Simplifying NO3 Dataframe

NO3 <- NO3 %>% dplyr::select(!c(nrow, dist, lon, lat))

NO3 <- NO3 %>% 
  pivot_longer(cols=c(2:217), names_to="NO3_date", names_prefix="NIT_")

NO3x <- NO3 
NO3x$NO3_date <- gsub("jan", "01-01-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("feb", "01-02-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("mar", "01-03-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("apr", "01-04-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("may", "01-05-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("jun", "01-06-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("jul", "01-07-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("aug", "01-08-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("sep", "01-09-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("oct", "01-10-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("nov", "01-11-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("dec", "01-12-20", NO3x$NO3_date)

NO3x$NO3_date <- format(as.Date(NO3x$NO3_date, format="%d-%m-%Y"),"%Y-%m-%d")
NO3x$NO3_date <- as.Date(NO3x$NO3_date)
NO3 <- NO3x
rm(NO3x)

40 PFF NO3 - Creating Time-Weighted Intervals of Exposures

NO3 <- NO3 %>% mutate(start=as.IDate(NO3_date))
NO3 <- NO3 %>% mutate(end=as.IDate(NO3_date + months(1) - days(1)))
NO3 <- as.data.table(NO3)
str(NO3)

## Classes 'data.table' and 'data.frame':   411480 obs. of  5 variables:
##  $ ID      : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ NO3_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value   : num  4 4.2 3.8 1.9 1.4 ...
##  $ start   : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end     : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

NO3_5yrWtedAvg <- intervalaverage(x=NO3,
                y=PFF_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

NO3_5yrWtedAvgx <- NO3_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
NO3_5yrWtedAvgx <- NO3_5yrWtedAvgx %>% rename("NO3"="value", "start_5yr"="start", "end_5yr"="end")
NO3_5yrWtedAvgx <- NO3_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(NO3_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   87626 obs. of  4 variables:
##  $ ID       : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ NO3      : num  2.16 2.17 2.17 2.17 2.17 ...
##  $ start_5yr: Date, format: "2001-08-06" "2001-09-05" ...
##  $ end_5yr  : Date, format: "2006-08-06" "2006-09-05" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to PFF

PFF <- left_join(PFF, NO3_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

41 PFF NH4 - Importing NH4 Datasets

Here I am importing the file which contains monthly NH4 level estimates by satellite at nearest lon/lat to PFF patient residential addresses. These are linked to the patient ID.

outfile1 <- here("PFF_fILD_2000_2017_NH4_2021_11_05.xlsx")
NH4 <- read_excel(outfile1)

42 PFF NH4 - Simplifying NH4 Dataframe

NH4 <- NH4 %>% dplyr::select(!c(nrow, dist, lon, lat))

NH4 <- NH4 %>% 
  pivot_longer(cols=c(2:217), names_to="NH4_date", names_prefix="NH4_")

NH4x <- NH4 
NH4x$NH4_date <- gsub("jan", "01-01-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("feb", "01-02-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("mar", "01-03-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("apr", "01-04-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("may", "01-05-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("jun", "01-06-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("jul", "01-07-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("aug", "01-08-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("sep", "01-09-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("oct", "01-10-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("nov", "01-11-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("dec", "01-12-20", NH4x$NH4_date)

NH4x$NH4_date <- format(as.Date(NH4x$NH4_date, format="%d-%m-%Y"),"%Y-%m-%d")
NH4x$NH4_date <- as.Date(NH4x$NH4_date)
NH4 <- NH4x
rm(NH4x)

43 PFF NH4 - Creating Time-Weighted Intervals of Exposures

NH4 <- NH4 %>% mutate(start=as.IDate(NH4_date))
NH4 <- NH4 %>% mutate(end=as.IDate(NH4_date + months(1) - days(1)))
NH4 <- as.data.table(NH4)
str(NH4)

## Classes 'data.table' and 'data.frame':   411480 obs. of  5 variables:
##  $ ID      : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ NH4_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value   : num  1.7 2.1 1.3 1 0.9 ...
##  $ start   : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end     : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

NH4_5yrWtedAvg <- intervalaverage(x=NH4,
                y=PFF_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

NH4_5yrWtedAvgx <- NH4_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
NH4_5yrWtedAvgx <- NH4_5yrWtedAvgx %>% rename("NH4"="value", "start_5yr"="start", "end_5yr"="end")
NH4_5yrWtedAvgx <- NH4_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(NH4_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   87626 obs. of  4 variables:
##  $ ID       : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ NH4      : num  1.25 1.25 1.25 1.26 1.27 ...
##  $ start_5yr: Date, format: "2001-08-06" "2001-09-05" ...
##  $ end_5yr  : Date, format: "2006-08-06" "2006-09-05" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to PFF

PFF <- left_join(PFF, NH4_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

44 PFF BC - Importing BC Datasets

Here I am importing the file which contains monthly BC level estimates by satellite at nearest lon/lat to PFF patient residential addresses. These are linked to the patient ID.

outfile1 <- here("PFF_fILD_2000_2017_BC_2021_11_05.xlsx")
BC <- read_excel(outfile1)

45 PFF BC - Simplifying BC Dataframe

BC <- BC %>% dplyr::select(!c(nrow, dist, lon, lat))

BC <- BC %>% 
  pivot_longer(cols=c(2:217), names_to="BC_date", names_prefix="BC_")

BCx <- BC 
BCx$BC_date <- gsub("jan", "01-01-20", BCx$BC_date)
BCx$BC_date <- gsub("feb", "01-02-20", BCx$BC_date)
BCx$BC_date <- gsub("mar", "01-03-20", BCx$BC_date)
BCx$BC_date <- gsub("apr", "01-04-20", BCx$BC_date)
BCx$BC_date <- gsub("may", "01-05-20", BCx$BC_date)
BCx$BC_date <- gsub("jun", "01-06-20", BCx$BC_date)
BCx$BC_date <- gsub("jul", "01-07-20", BCx$BC_date)
BCx$BC_date <- gsub("aug", "01-08-20", BCx$BC_date)
BCx$BC_date <- gsub("sep", "01-09-20", BCx$BC_date)
BCx$BC_date <- gsub("oct", "01-10-20", BCx$BC_date)
BCx$BC_date <- gsub("nov", "01-11-20", BCx$BC_date)
BCx$BC_date <- gsub("dec", "01-12-20", BCx$BC_date)

BCx$BC_date <- format(as.Date(BCx$BC_date, format="%d-%m-%Y"),"%Y-%m-%d")
BCx$BC_date <- as.Date(BCx$BC_date)
BC <- BCx
rm(BCx)

46 PFF BC - Creating Time-Weighted Intervals of Exposures

BC <- BC %>% mutate(start=as.IDate(BC_date))
BC <- BC %>% mutate(end=as.IDate(BC_date + months(1) - days(1)))
BC <- as.data.table(BC)
str(BC)

## Classes 'data.table' and 'data.frame':   411480 obs. of  5 variables:
##  $ ID     : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ BC_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  0.7 0.5 0.7 0.5 0.5 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

BC_5yrWtedAvg <- intervalaverage(x=BC,
                y=PFF_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

BC_5yrWtedAvgx <- BC_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
BC_5yrWtedAvgx <- BC_5yrWtedAvgx %>% rename("BC"="value", "start_5yr"="start", "end_5yr"="end")
BC_5yrWtedAvgx <- BC_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(BC_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   87626 obs. of  4 variables:
##  $ ID       : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ BC       : num  0.568 0.568 0.57 0.575 0.574 ...
##  $ start_5yr: Date, format: "2001-08-06" "2001-09-05" ...
##  $ end_5yr  : Date, format: "2006-08-06" "2006-09-05" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to PFF

PFF <- left_join(PFF, BC_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

47 PFF OM - Importing OM Datasets

Here I am importing the file which contains monthly OM level estimates by satellite at nearest lon/lat to PFF patient residential addresses. These are linked to the patient ID.

outfile1 <- here("PFF_fILD_2000_2017_OM_2021_11_05.xlsx")
OM <- read_excel(outfile1)

48 PFF OM - Simplifying OM Dataframe

OM <- OM %>% dplyr::select(!c(nrow, dist, lon, lat))

OM <- OM %>% 
  pivot_longer(cols=c(2:217), names_to="OM_date", names_prefix="OM_")

OMx <- OM 
OMx$OM_date <- gsub("jan", "01-01-20", OMx$OM_date)
OMx$OM_date <- gsub("feb", "01-02-20", OMx$OM_date)
OMx$OM_date <- gsub("mar", "01-03-20", OMx$OM_date)
OMx$OM_date <- gsub("apr", "01-04-20", OMx$OM_date)
OMx$OM_date <- gsub("may", "01-05-20", OMx$OM_date)
OMx$OM_date <- gsub("jun", "01-06-20", OMx$OM_date)
OMx$OM_date <- gsub("jul", "01-07-20", OMx$OM_date)
OMx$OM_date <- gsub("aug", "01-08-20", OMx$OM_date)
OMx$OM_date <- gsub("sep", "01-09-20", OMx$OM_date)
OMx$OM_date <- gsub("oct", "01-10-20", OMx$OM_date)
OMx$OM_date <- gsub("nov", "01-11-20", OMx$OM_date)
OMx$OM_date <- gsub("dec", "01-12-20", OMx$OM_date)

OMx$OM_date <- format(as.Date(OMx$OM_date, format="%d-%m-%Y"),"%Y-%m-%d")
OMx$OM_date <- as.Date(OMx$OM_date)
OM <- OMx
rm(OMx)

49 PFF OM - Creating Time-Weighted Intervals of Exposures

OM <- OM %>% mutate(start=as.IDate(OM_date))
OM <- OM %>% mutate(end=as.IDate(OM_date + months(1) - days(1)))
OM <- as.data.table(OM)
str(OM)

## Classes 'data.table' and 'data.frame':   411480 obs. of  5 variables:
##  $ ID     : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ OM_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  3.8 2.1 2.7 1.8 1.7 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

OM_5yrWtedAvg <- intervalaverage(x=OM,
                y=PFF_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

OM_5yrWtedAvgx <- OM_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
OM_5yrWtedAvgx <- OM_5yrWtedAvgx %>% rename("OM"="value", "start_5yr"="start", "end_5yr"="end")
OM_5yrWtedAvgx <- OM_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(OM_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   87626 obs. of  4 variables:
##  $ ID       : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ OM       : num  2.59 2.59 2.58 2.58 2.56 ...
##  $ start_5yr: Date, format: "2001-08-06" "2001-09-05" ...
##  $ end_5yr  : Date, format: "2006-08-06" "2006-09-05" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to PFF

PFF <- left_join(PFF, OM_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

50 PFF SS - Importing SS Datasets

Here I am importing the file which contains monthly SS level estimates by satellite at nearest lon/lat to PFF patient residential addresses. These are linked to the patient ID.

outfile1 <- here("PFF_fILD_2000_2017_SS_2021_11_05.xlsx")
SS <- read_excel(outfile1)

51 PFF SS - Simplifying SS Dataframe

SS <- SS %>% dplyr::select(!c(nrow, dist, lon, lat))

SS <- SS %>% 
  pivot_longer(cols=c(2:217), names_to="SS_date", names_prefix="SS_")

SSx <- SS 
SSx$SS_date <- gsub("jan", "01-01-20", SSx$SS_date)
SSx$SS_date <- gsub("feb", "01-02-20", SSx$SS_date)
SSx$SS_date <- gsub("mar", "01-03-20", SSx$SS_date)
SSx$SS_date <- gsub("apr", "01-04-20", SSx$SS_date)
SSx$SS_date <- gsub("may", "01-05-20", SSx$SS_date)
SSx$SS_date <- gsub("jun", "01-06-20", SSx$SS_date)
SSx$SS_date <- gsub("jul", "01-07-20", SSx$SS_date)
SSx$SS_date <- gsub("aug", "01-08-20", SSx$SS_date)
SSx$SS_date <- gsub("sep", "01-09-20", SSx$SS_date)
SSx$SS_date <- gsub("oct", "01-10-20", SSx$SS_date)
SSx$SS_date <- gsub("nov", "01-11-20", SSx$SS_date)
SSx$SS_date <- gsub("dec", "01-12-20", SSx$SS_date)

SSx$SS_date <- format(as.Date(SSx$SS_date, format="%d-%m-%Y"),"%Y-%m-%d")
SSx$SS_date <- as.Date(SSx$SS_date)
SS <- SSx
rm(SSx)

52 PFF SS - Creating Time-Weighted Intervals of Exposures

SS <- SS %>% mutate(start=as.IDate(SS_date))
SS <- SS %>% mutate(end=as.IDate(SS_date + months(1) - days(1)))
SS <- as.data.table(SS)
str(SS)

## Classes 'data.table' and 'data.frame':   411480 obs. of  5 variables:
##  $ ID     : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ SS_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  0.4 0.4 0.5 0 0.4 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

SS_5yrWtedAvg <- intervalaverage(x=SS,
                y=PFF_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

SS_5yrWtedAvgx <- SS_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
SS_5yrWtedAvgx <- SS_5yrWtedAvgx %>% rename("SS"="value", "start_5yr"="start", "end_5yr"="end")
SS_5yrWtedAvgx <- SS_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(SS_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   87626 obs. of  4 variables:
##  $ ID       : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ SS       : num  0.208 0.207 0.203 0.198 0.193 ...
##  $ start_5yr: Date, format: "2001-08-06" "2001-09-05" ...
##  $ end_5yr  : Date, format: "2006-08-06" "2006-09-05" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to PFF

PFF <- left_join(PFF, SS_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

53 PFF Soil - Importing Soil Datasets

Here I am importing the file which contains monthly Soil level estimates by satellite at nearest lon/lat to PFF patient residential addresses. These are linked to the patient ID.

outfile1 <- here("PFF_fILD_2000_2017_Soil_2021_11_05.xlsx")
Soil <- read_excel(outfile1)

54 PFF Soil - Simplifying Soil Dataframe

Soil <- Soil %>% dplyr::select(!c(nrow, dist, lon, lat))

Soil <- Soil %>% 
  pivot_longer(cols=c(2:217), names_to="Soil_date", names_prefix="soil_")

Soilx <- Soil 
Soilx$Soil_date <- gsub("jan", "01-01-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("feb", "01-02-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("mar", "01-03-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("apr", "01-04-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("may", "01-05-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("jun", "01-06-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("jul", "01-07-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("aug", "01-08-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("sep", "01-09-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("oct", "01-10-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("nov", "01-11-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("dec", "01-12-20", Soilx$Soil_date)

Soilx$Soil_date <- format(as.Date(Soilx$Soil_date, format="%d-%m-%Y"),"%Y-%m-%d")
Soilx$Soil_date <- as.Date(Soilx$Soil_date)
Soil <- Soilx
rm(Soilx)

55 PFF Soil - Creating Time-Weighted Intervals of Exposures

Soil <- Soil %>% mutate(start=as.IDate(Soil_date))
Soil <- Soil %>% mutate(end=as.IDate(Soil_date + months(1) - days(1)))
Soil <- as.data.table(Soil)
str(Soil)

## Classes 'data.table' and 'data.frame':   411480 obs. of  5 variables:
##  $ ID       : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ Soil_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value    : num  0.4 0.3 0.5 0.5 0.5 ...
##  $ start    : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end      : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

Soil_5yrWtedAvg <- intervalaverage(x=Soil,
                y=PFF_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

Soil_5yrWtedAvgx <- Soil_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
Soil_5yrWtedAvgx <- Soil_5yrWtedAvgx %>% rename("Soil"="value", "start_5yr"="start", "end_5yr"="end")
Soil_5yrWtedAvgx <- Soil_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(Soil_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   87626 obs. of  4 variables:
##  $ ID       : num  1 1 1 1 1 1 1 1 1 1 ...
##  $ Soil     : num  0.462 0.459 0.46 0.463 0.457 ...
##  $ start_5yr: Date, format: "2001-08-06" "2001-09-05" ...
##  $ end_5yr  : Date, format: "2006-08-06" "2006-09-05" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to PFF

PFF <- left_join(PFF, Soil_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

56 CARE-PF - Importing CARE-PF Datasets

Here I am importing the file which contains monthly PM level estimates by satellite at nearest lon/lat to CAREPF patient residential addresses. These are linked to the patient ID.

outfile1 <- here("CAREPF_fILD_2000_2018_PM25_2021_09_16.xlsx")
PM <- read_excel(outfile1)

Here I am importing the file which I used for my CIMD work that contains the baseline clinical and demographic data for patients who have CIMD

outfile2 <- here("CAREPF_fILDPts_BaselineData_2021_10_22.xlsx")
CARE <- read_excel(outfile2)

57 CARE-PF - Simplifying PM Dataframe

PM <- PM %>% dplyr::select(!c(nrow, dist, lon, lat))
colnames(PM)

##   [1] "ID"         "PM25_jan00" "PM25_feb00" "PM25_mar00" "PM25_apr00"
##   [6] "PM25_may00" "PM25_jun00" "PM25_jul00" "PM25_aug00" "PM25_sep00"
##  [11] "PM25_oct00" "PM25_nov00" "PM25_dec00" "PM25_jan01" "PM25_feb01"
##  [16] "PM25_mar01" "PM25_apr01" "PM25_may01" "PM25_jun01" "PM25_jul01"
##  [21] "PM25_aug01" "PM25_sep01" "PM25_oct01" "PM25_nov01" "PM25_dec01"
##  [26] "PM25_jan02" "PM25_feb02" "PM25_mar02" "PM25_apr02" "PM25_may02"
##  [31] "PM25_jun02" "PM25_jul02" "PM25_aug02" "PM25_sep02" "PM25_oct02"
##  [36] "PM25_nov02" "PM25_dec02" "PM25_jan03" "PM25_feb03" "PM25_mar03"
##  [41] "PM25_apr03" "PM25_may03" "PM25_jun03" "PM25_jul03" "PM25_aug03"
##  [46] "PM25_sep03" "PM25_oct03" "PM25_nov03" "PM25_dec03" "PM25_jan04"
##  [51] "PM25_feb04" "PM25_mar04" "PM25_apr04" "PM25_may04" "PM25_jun04"
##  [56] "PM25_jul04" "PM25_aug04" "PM25_sep04" "PM25_oct04" "PM25_nov04"
##  [61] "PM25_dec04" "PM25_jan05" "PM25_feb05" "PM25_mar05" "PM25_apr05"
##  [66] "PM25_may05" "PM25_jun05" "PM25_jul05" "PM25_aug05" "PM25_sep05"
##  [71] "PM25_oct05" "PM25_nov05" "PM25_dec05" "PM25_jan06" "PM25_feb06"
##  [76] "PM25_mar06" "PM25_apr06" "PM25_may06" "PM25_jun06" "PM25_jul06"
##  [81] "PM25_aug06" "PM25_sep06" "PM25_oct06" "PM25_nov06" "PM25_dec06"
##  [86] "PM25_jan07" "PM25_feb07" "PM25_mar07" "PM25_apr07" "PM25_may07"
##  [91] "PM25_jun07" "PM25_jul07" "PM25_aug07" "PM25_sep07" "PM25_oct07"
##  [96] "PM25_nov07" "PM25_dec07" "PM25_jan08" "PM25_feb08" "PM25_mar08"
## [101] "PM25_apr08" "PM25_may08" "PM25_jun08" "PM25_jul08" "PM25_aug08"
## [106] "PM25_sep08" "PM25_oct08" "PM25_nov08" "PM25_dec08" "PM25_jan09"
## [111] "PM25_feb09" "PM25_mar09" "PM25_apr09" "PM25_may09" "PM25_jun09"
## [116] "PM25_jul09" "PM25_aug09" "PM25_sep09" "PM25_oct09" "PM25_nov09"
## [121] "PM25_dec09" "PM25_jan10" "PM25_feb10" "PM25_mar10" "PM25_apr10"
## [126] "PM25_may10" "PM25_jun10" "PM25_jul10" "PM25_aug10" "PM25_sep10"
## [131] "PM25_oct10" "PM25_nov10" "PM25_dec10" "PM25_jan11" "PM25_feb11"
## [136] "PM25_mar11" "PM25_apr11" "PM25_may11" "PM25_jun11" "PM25_jul11"
## [141] "PM25_aug11" "PM25_sep11" "PM25_oct11" "PM25_nov11" "PM25_dec11"
## [146] "PM25_jan12" "PM25_feb12" "PM25_mar12" "PM25_apr12" "PM25_may12"
## [151] "PM25_jun12" "PM25_jul12" "PM25_aug12" "PM25_sep12" "PM25_oct12"
## [156] "PM25_nov12" "PM25_dec12" "PM25_jan13" "PM25_feb13" "PM25_mar13"
## [161] "PM25_apr13" "PM25_may13" "PM25_jun13" "PM25_jul13" "PM25_aug13"
## [166] "PM25_sep13" "PM25_oct13" "PM25_nov13" "PM25_dec13" "PM25_jan14"
## [171] "PM25_feb14" "PM25_mar14" "PM25_apr14" "PM25_may14" "PM25_jun14"
## [176] "PM25_jul14" "PM25_aug14" "PM25_sep14" "PM25_oct14" "PM25_nov14"
## [181] "PM25_dec14" "PM25_jan15" "PM25_feb15" "PM25_mar15" "PM25_apr15"
## [186] "PM25_may15" "PM25_jun15" "PM25_jul15" "PM25_aug15" "PM25_sep15"
## [191] "PM25_oct15" "PM25_nov15" "PM25_dec15" "PM25_jan16" "PM25_feb16"
## [196] "PM25_mar16" "PM25_apr16" "PM25_may16" "PM25_jun16" "PM25_jul16"
## [201] "PM25_aug16" "PM25_sep16" "PM25_oct16" "PM25_nov16" "PM25_dec16"
## [206] "PM25_jan17" "PM25_feb17" "PM25_mar17" "PM25_apr17" "PM25_may17"
## [211] "PM25_jun17" "PM25_jul17" "PM25_aug17" "PM25_sep17" "PM25_oct17"
## [216] "PM25_nov17" "PM25_dec17" "PM25_jan18" "PM25_feb18" "PM25_mar18"
## [221] "PM25_apr18" "PM25_may18" "PM25_jun18" "PM25_aug18" "PM25_sep18"
## [226] "PM25_oct18" "PM25_nov18" "PM25_dec18"

PM <- PM %>% 
  pivot_longer(cols=c(2:228), names_to="PM_date", names_prefix="PM25_")

PMx <- PM 
PMx$PM_date <- gsub("jan", "01-01-20", PMx$PM_date)
PMx$PM_date <- gsub("feb", "01-02-20", PMx$PM_date)
PMx$PM_date <- gsub("mar", "01-03-20", PMx$PM_date)
PMx$PM_date <- gsub("apr", "01-04-20", PMx$PM_date)
PMx$PM_date <- gsub("may", "01-05-20", PMx$PM_date)
PMx$PM_date <- gsub("jun", "01-06-20", PMx$PM_date)
PMx$PM_date <- gsub("jul", "01-07-20", PMx$PM_date)
PMx$PM_date <- gsub("aug", "01-08-20", PMx$PM_date)
PMx$PM_date <- gsub("sep", "01-09-20", PMx$PM_date)
PMx$PM_date <- gsub("oct", "01-10-20", PMx$PM_date)
PMx$PM_date <- gsub("nov", "01-11-20", PMx$PM_date)
PMx$PM_date <- gsub("dec", "01-12-20", PMx$PM_date)

PMx$PM_date <- format(as.Date(PMx$PM_date, format="%d-%m-%Y"),"%Y-%m-%d")
PMx$PM_date <- as.Date(PMx$PM_date)
PM <- PMx
rm(PMx)

58 CARE-PF - Simplifying CARE Dataframe

58.1 CARE-PF - Death/Transplant/Censor Date

Extracting year of diagnosis and year of death/transplant/censoring

#Start with the year of diagnosis
CARE <- CARE %>% 
  mutate(dx_yrmo = format(as.Date(CARE$dx_date, format="%Y-%m-%d"),"%Y-%m"))
CARE <- CARE %>% 
  mutate(dx_yr = format(as.Date(CARE$dx_date, format="%Y-%m-%d"),"%Y"))
CARE$dx_yr <- as.numeric(CARE$dx_yr)

#Then the year of death or lung transplant
CARE <- CARE %>% 
  mutate(deathORtx_date = if_else(!is.na(tx_date), tx_date, death_date))
CARE <- CARE %>% 
  mutate(deathORtx_yrmo = format(as.Date(CARE$deathORtx_date, format="%Y-%m-%d"),"%Y-%m"))

#Then the year the records were last updated (i.e. year of censoring)
CARE <- CARE %>% 
  mutate(DeathTxCensor_date = if_else(!is.na(deathORtx_date), deathORtx_date, last_updated))
CARE <- CARE %>% 
  mutate(censor_yrmo = format(as.Date(CARE$DeathTxCensor_date, format="%Y-%m-%d"),"%Y-%m"))

58.2 CARE-PF - Removing Unnecessary Columns and Correcting Dates

CARE <- CARE %>% dplyr::select(c(ID, site, age_dx, sex, smokeHx, dich_smoking, race, dich_Race, avg_s, dx_group, dx, dob, dx_date, death_date, tx_date, DeathTxCensor_date, last_updated, pft_date, fvc_pct, dlco_pct, pft_timefromdx, status, died, txed, deadORtx, time_DeathTxCensor, dx_yr))
CARE <- CARE %>% 
  mutate_at(c("dob","death_date", "last_updated", "tx_date", "dx_date", "pft_date", "DeathTxCensor_date"), as.Date)
str(CARE)

## tibble [3,389 × 27] (S3: tbl_df/tbl/data.frame)
##  $ ID                : num [1:3389] 10200257 10200069 10200266 10200336 10200229 ...
##  $ site              : num [1:3389] 102 102 102 102 102 102 102 102 102 102 ...
##  $ age_dx            : num [1:3389] 32.3 69.2 53.2 54.8 63.2 ...
##  $ sex               : chr [1:3389] "F" "M" "M" "M" ...
##  $ smokeHx           : chr [1:3389] "Former" "Former" "Never" "Never" ...
##  $ dich_smoking      : chr [1:3389] "Ever" "Ever" "Never" "Never" ...
##  $ race              : chr [1:3389] "W" "W" "W" "W" ...
##  $ dich_Race         : chr [1:3389] "White" "White" "White" "White" ...
##  $ avg_s             : num [1:3389] -0.797 -0.491 -0.881 -1.005 -1.005 ...
##  $ dx_group          : chr [1:3389] "CTD-ILD" "CTD-ILD" "CTD-ILD" "CTD-ILD" ...
##  $ dx                : chr [1:3389] "ANTISYNTHETASE" "RA_ILD" "UNDIFF_CTD_ILD" "SJOGRENS_ILD" ...
##  $ dob               : Date[1:3389], format: "1980-08-27" "1947-05-04" ...
##  $ dx_date           : Date[1:3389], format: "2012-12-15" "2016-07-14" ...
##  $ death_date        : Date[1:3389], format: NA "2017-07-12" ...
##  $ tx_date           : Date[1:3389], format: NA NA ...
##  $ DeathTxCensor_date: Date[1:3389], format: "2019-01-17" "2017-07-12" ...
##  $ last_updated      : Date[1:3389], format: "2019-01-17" "2017-07-12" ...
##  $ pft_date          : Date[1:3389], format: "2012-11-15" "2016-07-15" ...
##  $ fvc_pct           : num [1:3389] 79 105 97 78 80 63 62 65 96 82 ...
##  $ dlco_pct          : num [1:3389] 75 57 97 74 NA 44 66 49 88 64 ...
##  $ pft_timefromdx    : num [1:3389] -0.08214 0.00274 0 0 0.49555 ...
##  $ status            : num [1:3389] 0 1 0 0 1 0 1 0 0 0 ...
##  $ died              : chr [1:3389] "0" "1" "0" "0" ...
##  $ txed              : chr [1:3389] "0" "0" "0" "0" ...
##  $ deadORtx          : chr [1:3389] "0" "1" "0" "0" ...
##  $ time_DeathTxCensor: num [1:3389] 6.089 0.994 2.91 3.365 4.6 ...
##  $ dx_yr             : num [1:3389] 2012 2016 2017 2017 2014 ...

58.3 CARE-PF - Correcting Smoking

Need to correct smoking variables

CARE$smokeHx <- as.character(CARE$smokeHx) 
CARE <- CARE %>% mutate(smokeHx1=if_else(is.na(smokeHx), "Unknown", smokeHx))

#now need to make new dich_smoking category
CARE$dich_smoking <- as.character(CARE$dich_smoking) 
CARE <- CARE %>% mutate(dich_smoking1=if_else(is.na(dich_smoking),"Unknown", dich_smoking))

Now need to remove old smoking variables and rename new ones

CARE <- CARE %>% dplyr::select(-c(smokeHx, dich_smoking))
CARE <- CARE %>% rename(c("smokeHx"="smokeHx1", "dich_smoking"="dich_smoking1"))
CARE$smokeHx <- as.factor(CARE$smokeHx)
CARE$dich_smoking <- as.factor(CARE$dich_smoking)
CARE$smokeHx <- fct_relevel(CARE$smokeHx, c("Never","Former","Always","Unknown"))
CARE$dich_smoking <- fct_relevel(CARE$dich_smoking, c("Never","Ever","Unknown"))

58.4 CARE-PF - Correcting Factors

Need to correct other factor variables

CARE$sex <- fct_relevel(CARE$sex, c("M","F"))
CARE$race <- fct_relevel(CARE$race, c("W","B","A","N","U"))
CARE$dich_Race <- fct_relevel(CARE$dich_Race, c("White","Non-White"))
CARE$dx <- fct_relevel(CARE$dx, c("IPF"))
CARE$dx_group <- fct_relevel(CARE$dx_group, c("IPF"))
CARE <- CARE %>% mutate_at(c("status","died", "txed", "deadORtx", "site"), as.factor)

58.5 CARE-PF - Creating New Variables

58.5.1 CARE-PF - IPF vs Other Diagnosis

CARE <- CARE %>% mutate(dx_IPF=ifelse(dx=="IPF", "IPF", "not_IPF"))  
CARE$dx_IPF <- fct_relevel(CARE$dx_IPF, c("IPF"))

58.5.2 CARE-PF - Disadvantage Distribution

Creating this empirical cumulative distribution will allow us to combine the analyses of all three cohorts even though the measurements for disadvantage are different between the three.

plot(ecdf(CARE$avg_s))

CARE$disadv <- ecdf(CARE$avg_s)(CARE$avg_s)

59 CARE-PF - Modifying CARE Dataset

Take down CARE dataset to the 3389 patients with complete data

IDs <- as.data.table(unique(PM$ID))
IDs <- IDs %>% rename("ID"="V1")
CARE <- left_join(IDs, CARE, by="ID")

CARE <- CARE %>% mutate(days_DeathTxCensor=(time_DeathTxCensor*365.25))

Longest time_DeathTxCensor= 9546days

30-day longest time interval would be 9570 days

start <- seq(1, 9541, by = 30)
start

##   [1]    1   31   61   91  121  151  181  211  241  271  301  331  361  391  421
##  [16]  451  481  511  541  571  601  631  661  691  721  751  781  811  841  871
##  [31]  901  931  961  991 1021 1051 1081 1111 1141 1171 1201 1231 1261 1291 1321
##  [46] 1351 1381 1411 1441 1471 1501 1531 1561 1591 1621 1651 1681 1711 1741 1771
##  [61] 1801 1831 1861 1891 1921 1951 1981 2011 2041 2071 2101 2131 2161 2191 2221
##  [76] 2251 2281 2311 2341 2371 2401 2431 2461 2491 2521 2551 2581 2611 2641 2671
##  [91] 2701 2731 2761 2791 2821 2851 2881 2911 2941 2971 3001 3031 3061 3091 3121
## [106] 3151 3181 3211 3241 3271 3301 3331 3361 3391 3421 3451 3481 3511 3541 3571
## [121] 3601 3631 3661 3691 3721 3751 3781 3811 3841 3871 3901 3931 3961 3991 4021
## [136] 4051 4081 4111 4141 4171 4201 4231 4261 4291 4321 4351 4381 4411 4441 4471
## [151] 4501 4531 4561 4591 4621 4651 4681 4711 4741 4771 4801 4831 4861 4891 4921
## [166] 4951 4981 5011 5041 5071 5101 5131 5161 5191 5221 5251 5281 5311 5341 5371
## [181] 5401 5431 5461 5491 5521 5551 5581 5611 5641 5671 5701 5731 5761 5791 5821
## [196] 5851 5881 5911 5941 5971 6001 6031 6061 6091 6121 6151 6181 6211 6241 6271
## [211] 6301 6331 6361 6391 6421 6451 6481 6511 6541 6571 6601 6631 6661 6691 6721
## [226] 6751 6781 6811 6841 6871 6901 6931 6961 6991 7021 7051 7081 7111 7141 7171
## [241] 7201 7231 7261 7291 7321 7351 7381 7411 7441 7471 7501 7531 7561 7591 7621
## [256] 7651 7681 7711 7741 7771 7801 7831 7861 7891 7921 7951 7981 8011 8041 8071
## [271] 8101 8131 8161 8191 8221 8251 8281 8311 8341 8371 8401 8431 8461 8491 8521
## [286] 8551 8581 8611 8641 8671 8701 8731 8761 8791 8821 8851 8881 8911 8941 8971
## [301] 9001 9031 9061 9091 9121 9151 9181 9211 9241 9271 9301 9331 9361 9391 9421
## [316] 9451 9481 9511 9541

end <- seq(30, 9570, by = 30)
end

##   [1]   30   60   90  120  150  180  210  240  270  300  330  360  390  420  450
##  [16]  480  510  540  570  600  630  660  690  720  750  780  810  840  870  900
##  [31]  930  960  990 1020 1050 1080 1110 1140 1170 1200 1230 1260 1290 1320 1350
##  [46] 1380 1410 1440 1470 1500 1530 1560 1590 1620 1650 1680 1710 1740 1770 1800
##  [61] 1830 1860 1890 1920 1950 1980 2010 2040 2070 2100 2130 2160 2190 2220 2250
##  [76] 2280 2310 2340 2370 2400 2430 2460 2490 2520 2550 2580 2610 2640 2670 2700
##  [91] 2730 2760 2790 2820 2850 2880 2910 2940 2970 3000 3030 3060 3090 3120 3150
## [106] 3180 3210 3240 3270 3300 3330 3360 3390 3420 3450 3480 3510 3540 3570 3600
## [121] 3630 3660 3690 3720 3750 3780 3810 3840 3870 3900 3930 3960 3990 4020 4050
## [136] 4080 4110 4140 4170 4200 4230 4260 4290 4320 4350 4380 4410 4440 4470 4500
## [151] 4530 4560 4590 4620 4650 4680 4710 4740 4770 4800 4830 4860 4890 4920 4950
## [166] 4980 5010 5040 5070 5100 5130 5160 5190 5220 5250 5280 5310 5340 5370 5400
## [181] 5430 5460 5490 5520 5550 5580 5610 5640 5670 5700 5730 5760 5790 5820 5850
## [196] 5880 5910 5940 5970 6000 6030 6060 6090 6120 6150 6180 6210 6240 6270 6300
## [211] 6330 6360 6390 6420 6450 6480 6510 6540 6570 6600 6630 6660 6690 6720 6750
## [226] 6780 6810 6840 6870 6900 6930 6960 6990 7020 7050 7080 7110 7140 7170 7200
## [241] 7230 7260 7290 7320 7350 7380 7410 7440 7470 7500 7530 7560 7590 7620 7650
## [256] 7680 7710 7740 7770 7800 7830 7860 7890 7920 7950 7980 8010 8040 8070 8100
## [271] 8130 8160 8190 8220 8250 8280 8310 8340 8370 8400 8430 8460 8490 8520 8550
## [286] 8580 8610 8640 8670 8700 8730 8760 8790 8820 8850 8880 8910 8940 8970 9000
## [301] 9030 9060 9090 9120 9150 9180 9210 9240 9270 9300 9330 9360 9390 9420 9450
## [316] 9480 9510 9540 9570

Repeat the list of intervals 3389 times (number of patients in CARE)

start <- rep(start, times=3389)
end <- rep(end, times=3389)
intervals <- as.data.frame(cbind(start, end))

Add ID column to intervals

IDs <- rep(CARE$ID, each=319)
intervals <- as.data.frame(cbind(IDs, intervals))
intervals <- intervals %>% rename("ID"="IDs")

Join CARE and intervals

CARE <- left_join(intervals, CARE, by="ID")

Determine if event occurred during interval

CARE <- CARE %>% mutate(event=if_else((days_DeathTxCensor>=start & days_DeathTxCensor<=end), 1, 0))

Now will add date intervals for 5yr start and end times)

CARE <- CARE %>% mutate(end_5yr=(dx_date + days(end)))
CARE <- CARE %>% mutate(start_5yr=(end_5yr - days(1826)))

Now need to remove any rows where end_5yr is > date_DeathTxCensor (Amanda indicated need to remove rows where start_5yr > date_DeathTxCensor, but I don’t think this is correct because then we’d be assigning exposures to patients where they were alive/not transplanted for <5yrs for that exposure - need to double check with her)

CARE <- CARE %>% filter(!(end_5yr>DeathTxCensor_date & event!=1))

60 CARE-PF - Creating Time-Weighted Intervals of Exposures

PM <- PM %>% mutate(start=as.IDate(PM_date))
PM <- PM %>% mutate(end=as.IDate(PM_date + months(1) - days(1)))
PM <- as.data.table(PM)
str(PM)

## Classes 'data.table' and 'data.frame':   769303 obs. of  5 variables:
##  $ ID     : num  10100126 10100126 10100126 10100126 10100126 ...
##  $ PM_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  6 7.4 7.9 5.8 8.3 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

Creating a list of intervals we want to calculate exposures for:

CARE_intervals <- CARE %>% dplyr::select(ID, start_5yr, end_5yr)
CARE_intervals <- CARE_intervals %>% mutate_at(c("start_5yr", "end_5yr"), as.IDate)
CARE_intervals <- CARE_intervals %>% rename(c("start"="start_5yr", "end"="end_5yr"))
CARE_intervals <- as.data.table(CARE_intervals)
str(CARE_intervals)

## Classes 'data.table' and 'data.frame':   160215 obs. of  3 variables:
##  $ ID   : num  10100126 10100126 10100126 10100331 10100331 ...
##  $ start: IDate, format: "2012-09-10" "2012-10-10" ...
##  $ end  : IDate, format: "2017-09-10" "2017-10-10" ...
##  - attr(*, ".internal.selfref")=<externalptr>

PM_5yrWtedAvg <- intervalaverage(x=PM,
                y=CARE_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

PM_5yrWtedAvgx <- PM_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
PM_5yrWtedAvgx <- PM_5yrWtedAvgx %>% rename("PM"="value", "start_5yr"="start", "end_5yr"="end")
PM_5yrWtedAvgx <- PM_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(PM_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   160215 obs. of  4 variables:
##  $ ID       : num  10100001 10100001 10100001 10100001 10100001 ...
##  $ PM       : num  7.45 7.42 7.4 7.37 7.33 ...
##  $ start_5yr: Date, format: "2006-02-20" "2006-03-22" ...
##  $ end_5yr  : Date, format: "2011-02-20" "2011-03-22" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to CARE

CARE <- left_join(CARE, PM_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

61 CARE SO4 - Importing SO4 Datasets

Here I am importing the file which contains monthly SO4 level estimates by satellite at nearest lon/lat to CARE patient residential addresses. These are linked to the patient ID.

outfile1 <- here("CARE_fILD_2000_2017_SO4_2021_11_05.xlsx")
SO4 <- read_excel(outfile1)

62 CARE SO4 - Simplifying SO4 Dataframe

SO4 <- SO4 %>% dplyr::select(!c(nrow, dist, lon, lat))

SO4 <- SO4 %>% 
  pivot_longer(cols=c(2:217), names_to="SO4_date", names_prefix="SO4_")

SO4x <- SO4 
SO4x$SO4_date <- gsub("jan", "01-01-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("feb", "01-02-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("mar", "01-03-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("apr", "01-04-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("may", "01-05-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("jun", "01-06-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("jul", "01-07-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("aug", "01-08-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("sep", "01-09-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("oct", "01-10-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("nov", "01-11-20", SO4x$SO4_date)
SO4x$SO4_date <- gsub("dec", "01-12-20", SO4x$SO4_date)

SO4x$SO4_date <- format(as.Date(SO4x$SO4_date, format="%d-%m-%Y"),"%Y-%m-%d")
SO4x$SO4_date <- as.Date(SO4x$SO4_date)
SO4 <- SO4x
rm(SO4x)

63 CARE SO4 - Creating Time-Weighted Intervals of Exposures

SO4 <- SO4 %>% mutate(start=as.IDate(SO4_date))
SO4 <- SO4 %>% mutate(end=as.IDate(SO4_date + months(1) - days(1)))
SO4 <- as.data.table(SO4)
str(SO4)

## Classes 'data.table' and 'data.frame':   732024 obs. of  5 variables:
##  $ ID      : num  10200257 10200257 10200257 10200257 10200257 ...
##  $ SO4_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value   : num  1.3 0.4 0.2 0.5 0.6 ...
##  $ start   : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end     : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

SO4_5yrWtedAvg <- intervalaverage(x=SO4,
                y=CARE_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

SO4_5yrWtedAvgx <- SO4_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
SO4_5yrWtedAvgx <- SO4_5yrWtedAvgx %>% rename("SO4"="value", "start_5yr"="start", "end_5yr"="end")
SO4_5yrWtedAvgx <- SO4_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(SO4_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   160215 obs. of  4 variables:
##  $ ID       : num  10100001 10100001 10100001 10100001 10100001 ...
##  $ SO4      : num  2.03 2.01 2.01 1.99 1.97 ...
##  $ start_5yr: Date, format: "2006-02-20" "2006-03-22" ...
##  $ end_5yr  : Date, format: "2011-02-20" "2011-03-22" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to CARE

CARE <- left_join(CARE, SO4_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

64 CARE NO3 - Importing NO3 Datasets

Here I am importing the file which contains monthly NO3 level estimates by satellite at nearest lon/lat to CARE patient residential addresses. These are linked to the patient ID.

outfile1 <- here("CARE_fILD_2000_2017_NO3_2021_11_05.xlsx")
NO3 <- read_excel(outfile1)

65 CARE NO3 - Simplifying NO3 Dataframe

NO3 <- NO3 %>% dplyr::select(!c(nrow, dist, lon, lat))

NO3 <- NO3 %>% 
  pivot_longer(cols=c(2:217), names_to="NO3_date", names_prefix="NIT_")

NO3x <- NO3 
NO3x$NO3_date <- gsub("jan", "01-01-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("feb", "01-02-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("mar", "01-03-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("apr", "01-04-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("may", "01-05-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("jun", "01-06-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("jul", "01-07-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("aug", "01-08-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("sep", "01-09-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("oct", "01-10-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("nov", "01-11-20", NO3x$NO3_date)
NO3x$NO3_date <- gsub("dec", "01-12-20", NO3x$NO3_date)

NO3x$NO3_date <- format(as.Date(NO3x$NO3_date, format="%d-%m-%Y"),"%Y-%m-%d")
NO3x$NO3_date <- as.Date(NO3x$NO3_date)
NO3 <- NO3x
rm(NO3x)

66 CARE NO3 - Creating Time-Weighted Intervals of Exposures

NO3 <- NO3 %>% mutate(start=as.IDate(NO3_date))
NO3 <- NO3 %>% mutate(end=as.IDate(NO3_date + months(1) - days(1)))
NO3 <- as.data.table(NO3)
str(NO3)

## Classes 'data.table' and 'data.frame':   732024 obs. of  5 variables:
##  $ ID      : num  10200257 10200257 10200257 10200257 10200257 ...
##  $ NO3_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value   : num  0.9 0.1 0.1 0.8 0.6 ...
##  $ start   : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end     : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

NO3_5yrWtedAvg <- intervalaverage(x=NO3,
                y=CARE_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

NO3_5yrWtedAvgx <- NO3_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
NO3_5yrWtedAvgx <- NO3_5yrWtedAvgx %>% rename("NO3"="value", "start_5yr"="start", "end_5yr"="end")
NO3_5yrWtedAvgx <- NO3_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(NO3_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   160215 obs. of  4 variables:
##  $ ID       : num  10100001 10100001 10100001 10100001 10100001 ...
##  $ NO3      : num  0.889 0.888 0.885 0.885 0.887 ...
##  $ start_5yr: Date, format: "2006-02-20" "2006-03-22" ...
##  $ end_5yr  : Date, format: "2011-02-20" "2011-03-22" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to CARE

CARE <- left_join(CARE, NO3_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

67 CARE NH4 - Importing NH4 Datasets

Here I am importing the file which contains monthly NH4 level estimates by satellite at nearest lon/lat to CARE patient residential addresses. These are linked to the patient ID.

outfile1 <- here("CARE_fILD_2000_2017_NH4_2021_11_05.xlsx")
NH4 <- read_excel(outfile1)

68 CARE NH4 - Simplifying NH4 Dataframe

NH4 <- NH4 %>% dplyr::select(!c(nrow, dist, lon, lat))

NH4 <- NH4 %>% 
  pivot_longer(cols=c(2:217), names_to="NH4_date", names_prefix="NH4_")

NH4x <- NH4 
NH4x$NH4_date <- gsub("jan", "01-01-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("feb", "01-02-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("mar", "01-03-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("apr", "01-04-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("may", "01-05-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("jun", "01-06-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("jul", "01-07-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("aug", "01-08-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("sep", "01-09-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("oct", "01-10-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("nov", "01-11-20", NH4x$NH4_date)
NH4x$NH4_date <- gsub("dec", "01-12-20", NH4x$NH4_date)

NH4x$NH4_date <- format(as.Date(NH4x$NH4_date, format="%d-%m-%Y"),"%Y-%m-%d")
NH4x$NH4_date <- as.Date(NH4x$NH4_date)
NH4 <- NH4x
rm(NH4x)

69 CARE NH4 - Creating Time-Weighted Intervals of Exposures

NH4 <- NH4 %>% mutate(start=as.IDate(NH4_date))
NH4 <- NH4 %>% mutate(end=as.IDate(NH4_date + months(1) - days(1)))
NH4 <- as.data.table(NH4)
str(NH4)

## Classes 'data.table' and 'data.frame':   732024 obs. of  5 variables:
##  $ ID      : num  10200257 10200257 10200257 10200257 10200257 ...
##  $ NH4_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value   : num  0.8 0.4 0.2 0.3 0.4 ...
##  $ start   : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end     : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

NH4_5yrWtedAvg <- intervalaverage(x=NH4,
                y=CARE_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

NH4_5yrWtedAvgx <- NH4_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
NH4_5yrWtedAvgx <- NH4_5yrWtedAvgx %>% rename("NH4"="value", "start_5yr"="start", "end_5yr"="end")
NH4_5yrWtedAvgx <- NH4_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(NH4_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   160215 obs. of  4 variables:
##  $ ID       : num  10100001 10100001 10100001 10100001 10100001 ...
##  $ NH4      : num  0.89 0.883 0.88 0.871 0.861 ...
##  $ start_5yr: Date, format: "2006-02-20" "2006-03-22" ...
##  $ end_5yr  : Date, format: "2011-02-20" "2011-03-22" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to CARE

CARE <- left_join(CARE, NH4_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

70 CARE BC - Importing BC Datasets

Here I am importing the file which contains monthly BC level estimates by satellite at nearest lon/lat to CARE patient residential addresses. These are linked to the patient ID.

outfile1 <- here("CARE_fILD_2000_2017_BC_2021_11_05.xlsx")
BC <- read_excel(outfile1)

71 CARE BC - Simplifying BC Dataframe

BC <- BC %>% dplyr::select(!c(nrow, dist, lon, lat))

BC <- BC %>% 
  pivot_longer(cols=c(2:217), names_to="BC_date", names_prefix="BC_")

BCx <- BC 
BCx$BC_date <- gsub("jan", "01-01-20", BCx$BC_date)
BCx$BC_date <- gsub("feb", "01-02-20", BCx$BC_date)
BCx$BC_date <- gsub("mar", "01-03-20", BCx$BC_date)
BCx$BC_date <- gsub("apr", "01-04-20", BCx$BC_date)
BCx$BC_date <- gsub("may", "01-05-20", BCx$BC_date)
BCx$BC_date <- gsub("jun", "01-06-20", BCx$BC_date)
BCx$BC_date <- gsub("jul", "01-07-20", BCx$BC_date)
BCx$BC_date <- gsub("aug", "01-08-20", BCx$BC_date)
BCx$BC_date <- gsub("sep", "01-09-20", BCx$BC_date)
BCx$BC_date <- gsub("oct", "01-10-20", BCx$BC_date)
BCx$BC_date <- gsub("nov", "01-11-20", BCx$BC_date)
BCx$BC_date <- gsub("dec", "01-12-20", BCx$BC_date)

BCx$BC_date <- format(as.Date(BCx$BC_date, format="%d-%m-%Y"),"%Y-%m-%d")
BCx$BC_date <- as.Date(BCx$BC_date)
BC <- BCx
rm(BCx)

72 CARE BC - Creating Time-Weighted Intervals of Exposures

BC <- BC %>% mutate(start=as.IDate(BC_date))
BC <- BC %>% mutate(end=as.IDate(BC_date + months(1) - days(1)))
BC <- as.data.table(BC)
str(BC)

## Classes 'data.table' and 'data.frame':   732024 obs. of  5 variables:
##  $ ID     : num  10200257 10200257 10200257 10200257 10200257 ...
##  $ BC_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  0.4 0.3 0.2 0.2 0.2 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

BC_5yrWtedAvg <- intervalaverage(x=BC,
                y=CARE_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

BC_5yrWtedAvgx <- BC_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
BC_5yrWtedAvgx <- BC_5yrWtedAvgx %>% rename("BC"="value", "start_5yr"="start", "end_5yr"="end")
BC_5yrWtedAvgx <- BC_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(BC_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   160215 obs. of  4 variables:
##  $ ID       : num  10100001 10100001 10100001 10100001 10100001 ...
##  $ BC       : num  0.534 0.531 0.529 0.525 0.52 ...
##  $ start_5yr: Date, format: "2006-02-20" "2006-03-22" ...
##  $ end_5yr  : Date, format: "2011-02-20" "2011-03-22" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to CARE

CARE <- left_join(CARE, BC_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

73 CARE OM - Importing OM Datasets

Here I am importing the file which contains monthly OM level estimates by satellite at nearest lon/lat to CARE patient residential addresses. These are linked to the patient ID.

outfile1 <- here("CARE_fILD_2000_2017_OM_2021_11_05.xlsx")
OM <- read_excel(outfile1)

74 CARE OM - Simplifying OM Dataframe

OM <- OM %>% dplyr::select(!c(nrow, dist, lon, lat))

OM <- OM %>% 
  pivot_longer(cols=c(2:217), names_to="OM_date", names_prefix="OM_")

OMx <- OM 
OMx$OM_date <- gsub("jan", "01-01-20", OMx$OM_date)
OMx$OM_date <- gsub("feb", "01-02-20", OMx$OM_date)
OMx$OM_date <- gsub("mar", "01-03-20", OMx$OM_date)
OMx$OM_date <- gsub("apr", "01-04-20", OMx$OM_date)
OMx$OM_date <- gsub("may", "01-05-20", OMx$OM_date)
OMx$OM_date <- gsub("jun", "01-06-20", OMx$OM_date)
OMx$OM_date <- gsub("jul", "01-07-20", OMx$OM_date)
OMx$OM_date <- gsub("aug", "01-08-20", OMx$OM_date)
OMx$OM_date <- gsub("sep", "01-09-20", OMx$OM_date)
OMx$OM_date <- gsub("oct", "01-10-20", OMx$OM_date)
OMx$OM_date <- gsub("nov", "01-11-20", OMx$OM_date)
OMx$OM_date <- gsub("dec", "01-12-20", OMx$OM_date)

OMx$OM_date <- format(as.Date(OMx$OM_date, format="%d-%m-%Y"),"%Y-%m-%d")
OMx$OM_date <- as.Date(OMx$OM_date)
OM <- OMx
rm(OMx)

75 CARE OM - Creating Time-Weighted Intervals of Exposures

OM <- OM %>% mutate(start=as.IDate(OM_date))
OM <- OM %>% mutate(end=as.IDate(OM_date + months(1) - days(1)))
OM <- as.data.table(OM)
str(OM)

## Classes 'data.table' and 'data.frame':   732024 obs. of  5 variables:
##  $ ID     : num  10200257 10200257 10200257 10200257 10200257 ...
##  $ OM_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  2 1.1 0.4 0.7 1 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

OM_5yrWtedAvg <- intervalaverage(x=OM,
                y=CARE_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

OM_5yrWtedAvgx <- OM_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
OM_5yrWtedAvgx <- OM_5yrWtedAvgx %>% rename("OM"="value", "start_5yr"="start", "end_5yr"="end")
OM_5yrWtedAvgx <- OM_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(OM_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   160215 obs. of  4 variables:
##  $ ID       : num  10100001 10100001 10100001 10100001 10100001 ...
##  $ OM       : num  2.24 2.22 2.21 2.2 2.19 ...
##  $ start_5yr: Date, format: "2006-02-20" "2006-03-22" ...
##  $ end_5yr  : Date, format: "2011-02-20" "2011-03-22" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to CARE

CARE <- left_join(CARE, OM_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

76 CARE SS - Importing SS Datasets

Here I am importing the file which contains monthly SS level estimates by satellite at nearest lon/lat to CARE patient residential addresses. These are linked to the patient ID.

outfile1 <- here("CARE_fILD_2000_2017_SS_2021_11_05.xlsx")
SS <- read_excel(outfile1)

77 CARE SS - Simplifying SS Dataframe

SS <- SS %>% dplyr::select(!c(nrow, dist, lon, lat))

SS <- SS %>% 
  pivot_longer(cols=c(2:217), names_to="SS_date", names_prefix="SS_")

SSx <- SS 
SSx$SS_date <- gsub("jan", "01-01-20", SSx$SS_date)
SSx$SS_date <- gsub("feb", "01-02-20", SSx$SS_date)
SSx$SS_date <- gsub("mar", "01-03-20", SSx$SS_date)
SSx$SS_date <- gsub("apr", "01-04-20", SSx$SS_date)
SSx$SS_date <- gsub("may", "01-05-20", SSx$SS_date)
SSx$SS_date <- gsub("jun", "01-06-20", SSx$SS_date)
SSx$SS_date <- gsub("jul", "01-07-20", SSx$SS_date)
SSx$SS_date <- gsub("aug", "01-08-20", SSx$SS_date)
SSx$SS_date <- gsub("sep", "01-09-20", SSx$SS_date)
SSx$SS_date <- gsub("oct", "01-10-20", SSx$SS_date)
SSx$SS_date <- gsub("nov", "01-11-20", SSx$SS_date)
SSx$SS_date <- gsub("dec", "01-12-20", SSx$SS_date)

SSx$SS_date <- format(as.Date(SSx$SS_date, format="%d-%m-%Y"),"%Y-%m-%d")
SSx$SS_date <- as.Date(SSx$SS_date)
SS <- SSx
rm(SSx)

78 CARE SS - Creating Time-Weighted Intervals of Exposures

SS <- SS %>% mutate(start=as.IDate(SS_date))
SS <- SS %>% mutate(end=as.IDate(SS_date + months(1) - days(1)))
SS <- as.data.table(SS)
str(SS)

## Classes 'data.table' and 'data.frame':   732024 obs. of  5 variables:
##  $ ID     : num  10200257 10200257 10200257 10200257 10200257 ...
##  $ SS_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value  : num  0.1 0 0.2 0 0 ...
##  $ start  : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end    : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

SS_5yrWtedAvg <- intervalaverage(x=SS,
                y=CARE_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

SS_5yrWtedAvgx <- SS_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
SS_5yrWtedAvgx <- SS_5yrWtedAvgx %>% rename("SS"="value", "start_5yr"="start", "end_5yr"="end")
SS_5yrWtedAvgx <- SS_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(SS_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   160215 obs. of  4 variables:
##  $ ID       : num  10100001 10100001 10100001 10100001 10100001 ...
##  $ SS       : num  0.15 0.148 0.145 0.144 0.144 ...
##  $ start_5yr: Date, format: "2006-02-20" "2006-03-22" ...
##  $ end_5yr  : Date, format: "2011-02-20" "2011-03-22" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to CARE

CARE <- left_join(CARE, SS_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

79 CARE Soil - Importing Soil Datasets

Here I am importing the file which contains monthly Soil level estimates by satellite at nearest lon/lat to CARE patient residential addresses. These are linked to the patient ID.

outfile1 <- here("CARE_fILD_2000_2017_Soil_2021_11_05.xlsx")
Soil <- read_excel(outfile1)

80 CARE Soil - Simplifying Soil Dataframe

Soil <- Soil %>% dplyr::select(!c(nrow, dist, lon, lat))

Soil <- Soil %>% 
  pivot_longer(cols=c(2:217), names_to="Soil_date", names_prefix="soil_")

Soilx <- Soil 
Soilx$Soil_date <- gsub("jan", "01-01-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("feb", "01-02-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("mar", "01-03-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("apr", "01-04-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("may", "01-05-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("jun", "01-06-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("jul", "01-07-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("aug", "01-08-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("sep", "01-09-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("oct", "01-10-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("nov", "01-11-20", Soilx$Soil_date)
Soilx$Soil_date <- gsub("dec", "01-12-20", Soilx$Soil_date)

Soilx$Soil_date <- format(as.Date(Soilx$Soil_date, format="%d-%m-%Y"),"%Y-%m-%d")
Soilx$Soil_date <- as.Date(Soilx$Soil_date)
Soil <- Soilx
rm(Soilx)

81 CARE Soil - Creating Time-Weighted Intervals of Exposures

Soil <- Soil %>% mutate(start=as.IDate(Soil_date))
Soil <- Soil %>% mutate(end=as.IDate(Soil_date + months(1) - days(1)))
Soil <- as.data.table(Soil)
str(Soil)

## Classes 'data.table' and 'data.frame':   732024 obs. of  5 variables:
##  $ ID       : num  10200257 10200257 10200257 10200257 10200257 ...
##  $ Soil_date: Date, format: "2000-01-01" "2000-02-01" ...
##  $ value    : num  0.3 0.2 0 0.5 0.3 ...
##  $ start    : IDate, format: "2000-01-01" "2000-02-01" ...
##  $ end      : IDate, format: "2000-01-31" "2000-02-29" ...
##  - attr(*, ".internal.selfref")=<externalptr>

Soil_5yrWtedAvg <- intervalaverage(x=Soil,
                y=CARE_intervals,
                interval_vars=c("start","end"),
                value_vars=c("value"),
                group_vars="ID",
                required_percentage = 0.01)

So this produces a row for each interval for each patient

Select only the necessary columns, but will keep the original dataframe to interrogate data missingess if needed

Soil_5yrWtedAvgx <- Soil_5yrWtedAvg %>% dplyr::select(ID, value, start, end)
Soil_5yrWtedAvgx <- Soil_5yrWtedAvgx %>% rename("Soil"="value", "start_5yr"="start", "end_5yr"="end")
Soil_5yrWtedAvgx <- Soil_5yrWtedAvgx %>% mutate_at(c("start_5yr", "end_5yr"), as.Date)
str(Soil_5yrWtedAvgx)

## Classes 'data.table' and 'data.frame':   160215 obs. of  4 variables:
##  $ ID       : num  10100001 10100001 10100001 10100001 10100001 ...
##  $ Soil     : num  0.26 0.259 0.257 0.254 0.249 ...
##  $ start_5yr: Date, format: "2006-02-20" "2006-03-22" ...
##  $ end_5yr  : Date, format: "2011-02-20" "2011-03-22" ...
##  - attr(*, ".internal.selfref")=<externalptr> 
##  - attr(*, "sorted")= chr [1:3] "ID" "start_5yr" "end_5yr"

Join to CARE

CARE <- left_join(CARE, Soil_5yrWtedAvgx, by=c("ID", "start_5yr", "end_5yr"))

82 Remove unnecessary dataframes

rm(CARE_intervals, intervals, PFF_intervals, PM, PM_5yrWtedAvg, PM_5yrWtedAvgx, PM25, Simm_intervals, end, start, IDs, SO4, SO4_5yrWtedAvg, SO4_5yrWtedAvgx, NO3, NO3_5yrWtedAvg, NO3_5yrWtedAvgx, NH4, NH4_5yrWtedAvg, NH4_5yrWtedAvgx, BC, BC_5yrWtedAvg, BC_5yrWtedAvgx, OM, OM_5yrWtedAvg, OM_5yrWtedAvgx, SS, SS_5yrWtedAvg, SS_5yrWtedAvgx, Soil, Soil_5yrWtedAvg, Soil_5yrWtedAvgx)

83 Combine all into one dataframe

colnames(Simm)

##  [1] "ID"                 "start"              "end"               
##  [4] "dob"                "ADI_nat"            "death_date"        
##  [7] "last_updated"       "tx_date"            "dx_date"           
## [10] "consent_date"       "pft_date"           "fvc_pct"           
## [13] "dlco_pct"           "status"             "age_dx"            
## [16] "time_censoring"     "time_death"         "time_tx"           
## [19] "time_deathORtx"     "time_DeathTxCensor" "sex"               
## [22] "race"               "died"               "txed"              
## [25] "deadORtx"           "dx"                 "dich_Race"         
## [28] "dx_group"           "dx_yrmo"            "dx_yr"             
## [31] "DeathTxCensor_date" "censor_yrmo"        "smokeHx"           
## [34] "dich_smoking"       "dx_IPF"             "disadv"            
## [37] "days_DeathTxCensor" "event"              "end_5yr"           
## [40] "start_5yr"          "PM"                 "SO4"               
## [43] "NO3"                "NH4"                "BC"                
## [46] "OM"                 "SS"                 "Soil"

colnames(PFF)

##  [1] "ID"                 "start"              "end"               
##  [4] "site"               "age_dx"             "sex"               
##  [7] "race"               "dich_Race"          "pct_belowpoverty"  
## [10] "dx_group"           "dx"                 "dx_date"           
## [13] "death_date"         "tx_date"            "DeathTxCensor_date"
## [16] "censor_date"        "fvc_date"           "dlco_date"         
## [19] "fvc_pct"            "dlco_pct"           "status"            
## [22] "deadORtx"           "time_DeathTxCensor" "dx_yr"             
## [25] "smokeHx"            "dx_IPF"             "disadv"            
## [28] "days_DeathTxCensor" "event"              "end_5yr"           
## [31] "start_5yr"          "PM"                 "SO4"               
## [34] "NO3"                "NH4"                "BC"                
## [37] "OM"                 "SS"                 "Soil"

colnames(CARE)

##  [1] "ID"                 "start"              "end"               
##  [4] "site"               "age_dx"             "sex"               
##  [7] "race"               "dich_Race"          "avg_s"             
## [10] "dx_group"           "dx"                 "dob"               
## [13] "dx_date"            "death_date"         "tx_date"           
## [16] "DeathTxCensor_date" "last_updated"       "pft_date"          
## [19] "fvc_pct"            "dlco_pct"           "pft_timefromdx"    
## [22] "status"             "died"               "txed"              
## [25] "deadORtx"           "time_DeathTxCensor" "dx_yr"             
## [28] "smokeHx"            "dich_smoking"       "dx_IPF"            
## [31] "disadv"             "days_DeathTxCensor" "event"             
## [34] "end_5yr"            "start_5yr"          "PM"                
## [37] "SO4"                "NO3"                "NH4"               
## [40] "BC"                 "OM"                 "SS"                
## [43] "Soil"

Simm <- Simm %>% mutate(site=1)
Simm <- Simm %>% mutate(dlco_date=pft_date)
Simm <- Simm %>% rename(c("fvc_date"="pft_date"))
Simm <- Simm %>% dplyr::select(ID, start, end, site, age_dx, sex, race, dich_Race, smokeHx, dx_IPF, disadv, dx_group, dx, dx_yr, dx_date, death_date, tx_date, DeathTxCensor_date, last_updated, fvc_date, dlco_date, fvc_pct, dlco_pct, status, deadORtx, time_DeathTxCensor, days_DeathTxCensor, event, start_5yr, end_5yr, PM, SO4, NO3, NH4, BC, OM, SS, Soil)

PFF <- PFF %>% rename(c("last_updated"="censor_date"))
PFF <- PFF %>% dplyr::select(ID, start, end, site, age_dx, sex, race, dich_Race, smokeHx, dx_IPF, disadv, dx_group, dx, dx_yr, dx_date, death_date, tx_date, DeathTxCensor_date, last_updated, fvc_date, dlco_date, fvc_pct, dlco_pct, status, deadORtx, time_DeathTxCensor, days_DeathTxCensor, event, start_5yr, end_5yr, PM, SO4, NO3, NH4, BC, OM, SS, Soil)

CARE <- CARE %>% mutate(dlco_date=pft_date)
CARE <- CARE %>% rename(c("fvc_date"="pft_date"))
CARE <- CARE %>% dplyr::select(ID, start, end, site, age_dx, sex, race, dich_Race, smokeHx, dx_IPF, disadv, dx_group, dx, dx_yr, dx_date, death_date, tx_date, DeathTxCensor_date, last_updated, fvc_date, dlco_date, fvc_pct, dlco_pct, status, deadORtx, time_DeathTxCensor, days_DeathTxCensor, event, start_5yr, end_5yr, PM, SO4, NO3, NH4, BC, OM, SS, Soil)

84 Combining Cohorts

colnames(Simm)==colnames(PFF)

##  [1] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
## [16] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
## [31] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE

colnames(PFF)==colnames(CARE)

##  [1] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
## [16] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE
## [31] TRUE TRUE TRUE TRUE TRUE TRUE TRUE TRUE

84.1 Creating Cohort Column

Simm <- Simm %>% mutate(cohort="Simmons")
PFF <- PFF %>% mutate(cohort="PFF")
CARE <- CARE %>% mutate(cohort="CARE-PF")

84.2 Fixing Columns in PFF

Fixing sex

PFF <- PFF %>% mutate(sex1=if_else(sex=="Male", "M", "F"))
PFF <- PFF %>% dplyr::select(!sex)
PFF <- PFF %>% rename(c("sex"="sex1"))
PFF <- PFF %>% dplyr::select(ID, start, end, site, age_dx, sex, race, dich_Race, smokeHx, dx_IPF, disadv, dx_group, dx, dx_yr, dx_date, death_date, tx_date, DeathTxCensor_date, last_updated, fvc_date, dlco_date, fvc_pct, dlco_pct, status, deadORtx, time_DeathTxCensor, days_DeathTxCensor, event, start_5yr, end_5yr, PM, SO4, NO3, NH4, BC, OM, SS, Soil, cohort)

Fixing ID - some of the IDs are shared between PFF and Simm

#First want to check how many IDs are shared between Simm and PFF, Simm and CARE, and PFF and CARE
intersect(Simm$ID, PFF$ID)

## [1] 1097 1405

intersect(Simm$ID, CARE$ID)

## numeric(0)

intersect(CARE$ID, PFF$ID)

## numeric(0)

PFF$ID <- paste0(2000, PFF$ID)
intersect(Simm$ID, PFF$ID)

## character(0)

No overlaps between Simm/CARE or PFF/CARE, but two with Simm/PFF, so add “2000” to front of PFF IDs

84.3 Joining rows

All <- rbind(Simm, PFF)

## Warning in `[<-.factor`(`*tmp*`, ri, value = c(0, 0, 0, 0, 0, 0, 0, 0, 0, :
## invalid factor level, NA generated

All <- rbind(All, CARE)
str(All)

## 'data.frame':    337348 obs. of  39 variables:
##  $ ID                : chr  "1097" "1097" "1097" "1097" ...
##  $ start             : num  1 31 61 91 121 151 181 211 241 271 ...
##  $ end               : num  30 60 90 120 150 180 210 240 270 300 ...
##  $ site              : chr  "1" "1" "1" "1" ...
##  $ age_dx            : num  64.4 64.4 64.4 64.4 64.4 ...
##  $ sex               : Factor w/ 2 levels "M","F": 2 2 2 2 2 2 2 2 2 2 ...
##  $ race              : Factor w/ 7 levels "W","B","A","N",..: 2 2 2 2 2 2 2 2 2 2 ...
##  $ dich_Race         : Factor w/ 2 levels "White","Non-White": 2 2 2 2 2 2 2 2 2 2 ...
##  $ smokeHx           : Factor w/ 5 levels "Never","Former",..: 4 4 4 4 4 4 4 4 4 4 ...
##  $ dx_IPF            : Factor w/ 2 levels "IPF","not_IPF": 2 2 2 2 2 2 2 2 2 2 ...
##  $ disadv            : num  0.579 0.579 0.579 0.579 0.579 ...
##  $ dx_group          : Factor w/ 7 levels "IPF","CTD-ILD",..: 2 2 2 2 2 2 2 2 2 2 ...
##  $ dx                : Factor w/ 39 levels "IPF","AIP","AMYLOIDOSIS",..: 30 30 30 30 30 30 30 30 30 30 ...
##  $ dx_yr             : num  2002 2002 2002 2002 2002 ...
##  $ dx_date           : Date, format: "2002-01-24" "2002-01-24" ...
##  $ death_date        : Date, format: "2008-12-09" "2008-12-09" ...
##  $ tx_date           : Date, format: NA NA ...
##  $ DeathTxCensor_date: Date, format: "2008-12-09" "2008-12-09" ...
##  $ last_updated      : Date, format: "2021-01-27" "2021-01-27" ...
##  $ fvc_date          : Date, format: "2002-01-18" "2002-01-18" ...
##  $ dlco_date         : Date, format: "2002-01-18" "2002-01-18" ...
##  $ fvc_pct           : num  31.9 31.9 31.9 31.9 31.9 ...
##  $ dlco_pct          : num  NA NA NA NA NA NA NA NA NA NA ...
##  $ status            : Factor w/ 3 levels "0","1","2": 2 2 2 2 2 2 2 2 2 2 ...
##  $ deadORtx          : Factor w/ 2 levels "0","1": 2 2 2 2 2 2 2 2 2 2 ...
##  $ time_DeathTxCensor: num  6.87 6.87 6.87 6.87 6.87 ...
##  $ days_DeathTxCensor: num  2511 2511 2511 2511 2511 ...
##  $ event             : num  0 0 0 0 0 0 0 0 0 0 ...
##  $ start_5yr         : Date, format: "1997-02-23" "1997-03-25" ...
##  $ end_5yr           : Date, format: "2002-02-23" "2002-03-25" ...
##  $ PM                : num  12.5 12.4 12.3 12.3 12.5 ...
##  $ SO4               : num  4.41 4.37 4.35 4.33 4.41 ...
##  $ NO3               : num  1.4 1.41 1.4 1.38 1.36 ...
##  $ NH4               : num  1.7 1.69 1.68 1.67 1.69 ...
##  $ BC                : num  0.774 0.768 0.762 0.753 0.763 ...
##  $ OM                : num  2.9 2.89 2.88 2.85 2.9 ...
##  $ SS                : num  0.241 0.248 0.242 0.248 0.249 ...
##  $ Soil              : num  0.521 0.513 0.511 0.511 0.515 ...
##  $ cohort            : chr  "Simmons" "Simmons" "Simmons" "Simmons" ...

84.4 Correct Site and Cohort to Factors

All$cohort <- as.factor(All$cohort)
All$site <- as.factor(All$site)
All$event <- as.factor(All$event)

85 Export File with All PM2.5 and Constituent Time Frames Matched

write_xlsx(All, path="CombinedCohorts_fILD_AllConstituentsMatched_2023_01_02.xlsx")

86 Summary Data

86.1 Summary of Time-Varying PM2.5 and Constituent Levels

86.1.1 Simmons

Simm_medPM <- Simm %>% group_by(ID) %>% summarise(Mean=mean(PM, na.rm=T), Max=max(PM, na.rm=T), Min=min(PM, na.rm=T), Median=median(PM, na.rm=T), StdDev=sd(PM, na.rm=T), Q1=quantile(PM, 0.25, na.rm=T), Q3=quantile(PM, 0.75, na.rm=T))
Simm_medPM %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1  10.7  17.3  3.57   10.4  0.559  8.40  13.0

Simm_medSO4 <- Simm %>% group_by(ID) %>% summarise(Mean=mean(SO4, na.rm=T), Max=max(SO4, na.rm=T), Min=min(SO4, na.rm=T), Median=median(SO4, na.rm=T), StdDev=sd(SO4, na.rm=T), Q1=quantile(SO4, 0.25, na.rm=T), Q3=quantile(SO4, 0.75, na.rm=T))
Simm_medSO4 %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1  3.23  6.70 0.516   2.91  0.353  1.88  4.67

Simm_medNO3 <- Simm %>% group_by(ID) %>% summarise(Mean=mean(NO3, na.rm=T), Max=max(NO3, na.rm=T), Min=min(NO3, na.rm=T), Median=median(NO3, na.rm=T), StdDev=sd(NO3, na.rm=T), Q1=quantile(NO3, 0.25, na.rm=T), Q3=quantile(NO3, 0.75, na.rm=T))
Simm_medNO3 %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1  1.05  3.92 0.211   1.02 0.0518 0.825  1.25

Simm_medNH4 <- Simm %>% group_by(ID) %>% summarise(Mean=mean(NH4, na.rm=T), Max=max(NH4, na.rm=T), Min=min(NH4, na.rm=T), Median=median(NH4, na.rm=T), StdDev=sd(NH4, na.rm=T), Q1=quantile(NH4, 0.25, na.rm=T), Q3=quantile(NH4, 0.75, na.rm=T))
Simm_medNH4 %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max    Min Median StdDev    Q1    Q3
##   <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1  1.21  2.61 0.0361   1.14  0.145 0.663  1.77

Simm_medBC <- Simm %>% group_by(ID) %>% summarise(Mean=mean(BC, na.rm=T), Max=max(BC, na.rm=T), Min=min(BC, na.rm=T), Median=median(BC, na.rm=T), StdDev=sd(BC, na.rm=T), Q1=quantile(BC, 0.25, na.rm=T), Q3=quantile(BC, 0.75, na.rm=T))
Simm_medBC %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.842  1.88 0.240  0.829 0.0373 0.686  1.00

Simm_medOM <- Simm %>% group_by(ID) %>% summarise(Mean=mean(OM, na.rm=T), Max=max(OM, na.rm=T), Min=min(OM, na.rm=T), Median=median(OM, na.rm=T), StdDev=sd(OM, na.rm=T), Q1=quantile(OM, 0.25, na.rm=T), Q3=quantile(OM, 0.75, na.rm=T))
Simm_medOM %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1  3.22  6.81  1.11   3.16  0.107  2.67  3.66

Simm_medSS <- Simm %>% group_by(ID) %>% summarise(Mean=mean(SS, na.rm=T), Max=max(SS, na.rm=T), Min=min(SS, na.rm=T), Median=median(SS, na.rm=T), StdDev=sd(SS, na.rm=T), Q1=quantile(SS, 0.25, na.rm=T), Q3=quantile(SS, 0.75, na.rm=T))
Simm_medSS %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.206  1.51     0  0.176 0.0233 0.128 0.271

Simm_medSoil <- Simm %>% group_by(ID) %>% summarise(Mean=mean(Soil, na.rm=T), Max=max(Soil, na.rm=T), Min=min(Soil, na.rm=T), Median=median(Soil, na.rm=T), StdDev=sd(Soil, na.rm=T), Q1=quantile(Soil, 0.25, na.rm=T), Q3=quantile(Soil, 0.75, na.rm=T))
Simm_medSoil %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.482  1.36 0.123  0.487 0.0172 0.369 0.558

86.1.2 PFF

PFF_medPM <- PFF %>% group_by(ID) %>% summarise(Mean=mean(PM, na.rm=T), Max=max(PM, na.rm=T), Min=min(PM, na.rm=T), Median=median(PM, na.rm=T), StdDev=sd(PM, na.rm=T), Q1=quantile(PM, 0.25, na.rm=T), Q3=quantile(PM, 0.75, na.rm=T))

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(PM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(PM, na.rm = T): no non-missing arguments to min; returning Inf

PFF_medPM %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1  8.61  17.3  2.47   8.55  0.203  7.33  9.88

PFF_medPM <- inner_join(PFF_medPM, PFF, by="ID")
PFF_medPM %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 41 × 8
##    site   Mean   Max   Min Median StdDev    Q1    Q3
##    <fct> <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
##  1 01R    6.15 10.4   3.84   5.91 0.104   5.21  6.71
##  2 02R    8.09 12.6   3.70   8.12 0.159   6.53 10.0 
##  3 03R    6.09  9.37  2.53   6.41 0.110   4.16  7.79
##  4 04R   10.4  16.4   6.91   9.64 0.139   9.05 11.4 
##  5 05R   10.8  15.0   8.05  10.6  0.133   9.61 11.8 
##  6 06R    9.35 14.2   4.37   9.34 0.217   8.17 10.4 
##  7 07R    8.61 12.0   4.54   8.54 0.177   7.74  9.70
##  8 09R    8.03 14.8   5.30   7.63 0.152   6.56  8.92
##  9 10R    7.29  8.63  6.15   7.24 0.0666  6.96  7.69
## 10 11R    9.38 15.8   7.06   9.14 0.219   8.43 10.1 
## # … with 31 more rows

PFF_medSO4 <- PFF %>% group_by(ID) %>% summarise(Mean=mean(SO4, na.rm=T), Max=max(SO4, na.rm=T), Min=min(SO4, na.rm=T), Median=median(SO4, na.rm=T), StdDev=sd(SO4, na.rm=T), Q1=quantile(SO4, 0.25, na.rm=T), Q3=quantile(SO4, 0.75, na.rm=T))

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SO4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SO4, na.rm = T): no non-missing arguments to min; returning Inf

PFF_medSO4 %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1  1.59  5.16 0.236   1.55  0.119  1.12  2.06

PFF_medSO4 <- inner_join(PFF_medSO4, PFF, by="ID")
PFF_medSO4 %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 41 × 8
##    site   Mean   Max   Min Median StdDev    Q1    Q3
##    <fct> <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
##  1 01R   0.496  1.18 0.265  0.466 0.0204 0.360 0.566
##  2 02R   0.671  1.09 0.290  0.708 0.0142 0.574 0.767
##  3 03R   0.643  1.51 0.248  0.618 0.0260 0.487 0.815
##  4 04R   1.94   3.96 1.13   1.62  0.0614 1.42  2.22 
##  5 05R   2.64   4.93 1.37   2.50  0.106  1.88  3.57 
##  6 06R   2.24   5.03 0.749  1.99  0.134  1.70  2.57 
##  7 07R   1.65   3.44 0.729  1.56  0.0639 1.31  1.97 
##  8 09R   1.47   4.56 0.723  1.22  0.103  1.02  1.69 
##  9 10R   1.16   1.98 0.760  1.04  0.0675 0.897 1.43 
## 10 11R   1.90   5.16 1.09   1.79  0.108  1.51  2.18 
## # … with 31 more rows

PFF_medNO3 <- PFF %>% group_by(ID) %>% summarise(Mean=mean(NO3, na.rm=T), Max=max(NO3, na.rm=T), Min=min(NO3, na.rm=T), Median=median(NO3, na.rm=T), StdDev=sd(NO3, na.rm=T), Q1=quantile(NO3, 0.25, na.rm=T), Q3=quantile(NO3, 0.75, na.rm=T))

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NO3, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NO3, na.rm = T): no non-missing arguments to min; returning Inf

PFF_medNO3 %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max    Min Median StdDev    Q1    Q3
##   <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.987  4.76 0.0287  0.889 0.0476 0.505  1.42

PFF_medNO3 <- inner_join(PFF_medNO3, PFF, by="ID")
PFF_medNO3 %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 41 × 8
##    site   Mean   Max    Min Median  StdDev    Q1    Q3
##    <fct> <dbl> <dbl>  <dbl>  <dbl>   <dbl> <dbl> <dbl>
##  1 01R   0.405 0.763 0.0715  0.411 0.00859 0.323 0.522
##  2 02R   1.10  2.88  0.112   1.13  0.0279  0.736 1.41 
##  3 03R   0.755 1.74  0.0734  0.675 0.0316  0.272 1.30 
##  4 04R   1.83  3.60  0.372   1.80  0.0432  1.62  2.09 
##  5 05R   1.26  1.81  0.697   1.25  0.0303  1.10  1.38 
##  6 06R   0.748 1.76  0.134   0.702 0.0232  0.529 0.925
##  7 07R   1.45  2.06  0.345   1.44  0.0289  1.31  1.66 
##  8 09R   0.858 2.29  0.478   0.703 0.0199  0.608 1.05 
##  9 10R   0.340 0.542 0.255   0.334 0.00550 0.303 0.359
## 10 11R   0.416 0.778 0.220   0.438 0.0111  0.312 0.507
## # … with 31 more rows

PFF_medNH4 <- PFF %>% group_by(ID) %>% summarise(Mean=mean(NH4, na.rm=T), Max=max(NH4, na.rm=T), Min=min(NH4, na.rm=T), Median=median(NH4, na.rm=T), StdDev=sd(NH4, na.rm=T), Q1=quantile(NH4, 0.25, na.rm=T), Q3=quantile(NH4, 0.75, na.rm=T))

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(NH4, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(NH4, na.rm = T): no non-missing arguments to min; returning Inf

PFF_medNH4 %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max    Min Median StdDev    Q1    Q3
##   <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.574  2.48 0.0145  0.510 0.0610 0.309 0.806

PFF_medNH4 <- inner_join(PFF_medNH4, PFF, by="ID")
PFF_medNH4 %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 41 × 8
##    site   Mean   Max    Min Median StdDev     Q1    Q3
##    <fct> <dbl> <dbl>  <dbl>  <dbl>  <dbl>  <dbl> <dbl>
##  1 01R   0.213 0.652 0.0420 0.158  0.0161 0.112  0.312
##  2 02R   0.338 1.20  0.0394 0.303  0.0121 0.220  0.483
##  3 03R   0.277 0.635 0.0145 0.280  0.0168 0.0982 0.425
##  4 04R   0.934 2.17  0.0856 0.721  0.0485 0.566  1.25 
##  5 05R   1.01  2.04  0.368  0.899  0.0538 0.650  1.47 
##  6 06R   0.711 2.02  0.195  0.597  0.0565 0.434  0.917
##  7 07R   0.721 1.72  0.115  0.686  0.0382 0.507  0.901
##  8 09R   0.562 2.13  0.134  0.433  0.0467 0.296  0.671
##  9 10R   0.130 0.439 0.0347 0.0903 0.0168 0.0587 0.184
## 10 11R   0.398 1.55  0.0905 0.364  0.0434 0.266  0.509
## # … with 31 more rows

PFF_medBC <- PFF %>% group_by(ID) %>% summarise(Mean=mean(BC, na.rm=T), Max=max(BC, na.rm=T), Min=min(BC, na.rm=T), Median=median(BC, na.rm=T), StdDev=sd(BC, na.rm=T), Q1=quantile(BC, 0.25, na.rm=T), Q3=quantile(BC, 0.75, na.rm=T))

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(BC, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(BC, na.rm = T): no non-missing arguments to min; returning Inf

PFF_medBC %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max    Min Median StdDev    Q1    Q3
##   <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.657  2.54 0.0726  0.640 0.0378 0.498 0.806

PFF_medBC <- inner_join(PFF_medBC, PFF, by="ID")
PFF_medBC %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 41 × 8
##    site   Mean   Max   Min Median StdDev    Q1    Q3
##    <fct> <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
##  1 01R   0.464 1.07  0.214  0.383 0.0257 0.308 0.583
##  2 02R   0.517 0.943 0.161  0.521 0.0168 0.372 0.679
##  3 03R   0.603 1.22  0.133  0.489 0.0203 0.232 0.988
##  4 04R   0.729 1.09  0.508  0.696 0.0102 0.619 0.803
##  5 05R   0.809 1.28  0.588  0.778 0.0212 0.672 0.889
##  6 06R   0.749 1.11  0.344  0.745 0.0287 0.627 0.910
##  7 07R   0.610 0.940 0.317  0.587 0.0191 0.533 0.711
##  8 09R   0.622 1.46  0.335  0.480 0.0484 0.419 0.776
##  9 10R   0.378 0.544 0.280  0.348 0.0186 0.314 0.455
## 10 11R   0.831 1.65  0.533  0.809 0.0408 0.699 0.917
## # … with 31 more rows

PFF_medOM <- PFF %>% group_by(ID) %>% summarise(Mean=mean(OM, na.rm=T), Max=max(OM, na.rm=T), Min=min(OM, na.rm=T), Median=median(OM, na.rm=T), StdDev=sd(OM, na.rm=T), Q1=quantile(OM, 0.25, na.rm=T), Q3=quantile(OM, 0.75, na.rm=T))

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(OM, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(OM, na.rm = T): no non-missing arguments to min; returning Inf

PFF_medOM %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1  2.92  8.32 0.504   2.78 0.0912  2.26  3.51

PFF_medOM <- inner_join(PFF_medOM, PFF, by="ID")
PFF_medOM %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 41 × 8
##    site   Mean   Max   Min Median StdDev    Q1    Q3
##    <fct> <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
##  1 01R    2.55  5.66 1.12    2.03 0.0806  1.52  3.75
##  2 02R    2.93  5.54 0.936   3.00 0.107   2.05  3.82
##  3 03R    3.14  7.28 0.695   2.71 0.199   1.10  5.02
##  4 04R    2.74  4.03 2.25    2.60 0.0363  2.47  2.98
##  5 05R    3.39  5.08 2.45    3.06 0.0371  2.71  4.23
##  6 06R    3.25  5.09 1.51    3.17 0.0464  2.81  3.56
##  7 07R    2.88  4.35 1.55    2.73 0.0667  2.43  3.31
##  8 09R    2.87  6.02 1.69    2.21 0.0782  1.87  4.01
##  9 10R    1.63  2.87 1.14    1.51 0.0398  1.30  1.85
## 10 11R    3.98  6.44 2.97    3.99 0.0740  3.58  4.35
## # … with 31 more rows

PFF_medSS <- PFF %>% group_by(ID) %>% summarise(Mean=mean(SS, na.rm=T), Max=max(SS, na.rm=T), Min=min(SS, na.rm=T), Median=median(SS, na.rm=T), StdDev=sd(SS, na.rm=T), Q1=quantile(SS, 0.25, na.rm=T), Q3=quantile(SS, 0.75, na.rm=T))

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(SS, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(SS, na.rm = T): no non-missing arguments to min; returning Inf

PFF_medSS %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.371  2.71     0  0.245 0.0216 0.173 0.416

PFF_medSS <- inner_join(PFF_medSS, PFF, by="ID")
PFF_medSS %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 41 × 8
##    site   Mean   Max    Min Median  StdDev     Q1    Q3
##    <fct> <dbl> <dbl>  <dbl>  <dbl>   <dbl>  <dbl> <dbl>
##  1 01R   0.314 1.23  0.0742  0.241 0.0290  0.170  0.299
##  2 02R   1.47  2.71  0.0531  1.56  0.0291  0.799  2.13 
##  3 03R   0.168 1.57  0       0.112 0.0294  0.0185 0.207
##  4 04R   0.264 0.835 0.110   0.228 0.0155  0.191  0.266
##  5 05R   0.131 0.230 0.0703  0.122 0.00643 0.0911 0.173
##  6 06R   0.135 0.332 0.0696  0.124 0.00785 0.0994 0.160
##  7 07R   0.243 0.877 0.138   0.221 0.0120  0.189  0.267
##  8 09R   0.367 0.611 0.189   0.353 0.0129  0.284  0.425
##  9 10R   1.08  1.66  0.593   1.06  0.0344  0.925  1.25 
## 10 11R   0.268 0.688 0.127   0.236 0.00976 0.206  0.289
## # … with 31 more rows

PFF_medSoil <- PFF %>% group_by(ID) %>% summarise(Mean=mean(Soil, na.rm=T), Max=max(Soil, na.rm=T), Min=min(Soil, na.rm=T), Median=median(Soil, na.rm=T), StdDev=sd(Soil, na.rm=T), Q1=quantile(Soil, 0.25, na.rm=T), Q3=quantile(Soil, 0.75, na.rm=T))

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in max(Soil, na.rm = T): no non-missing arguments to max; returning -Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

## Warning in min(Soil, na.rm = T): no non-missing arguments to min; returning Inf

PFF_medSoil %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max    Min Median StdDev    Q1    Q3
##   <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.616  2.85 0.0621  0.494 0.0281 0.337 0.770

PFF_medSoil <- inner_join(PFF_medSoil, PFF, by="ID")
PFF_medSoil %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 41 × 8
##    site   Mean   Max    Min Median  StdDev    Q1    Q3
##    <fct> <dbl> <dbl>  <dbl>  <dbl>   <dbl> <dbl> <dbl>
##  1 01R   0.256 0.644 0.0829  0.200 0.0148  0.126 0.379
##  2 02R   0.405 1.01  0.0989  0.402 0.00942 0.290 0.508
##  3 03R   0.923 1.59  0.0839  0.844 0.0215  0.542 1.35 
##  4 04R   0.541 0.801 0.401   0.521 0.00716 0.484 0.578
##  5 05R   0.515 0.704 0.277   0.487 0.00740 0.429 0.649
##  6 06R   0.542 0.910 0.0686  0.533 0.0111  0.476 0.639
##  7 07R   0.525 1.18  0.188   0.440 0.0223  0.351 0.641
##  8 09R   0.282 0.719 0.0994  0.206 0.00833 0.155 0.403
##  9 10R   1.04  1.42  0.578   1.05  0.0290  0.871 1.20 
## 10 11R   0.770 1.11  0.501   0.758 0.0126  0.691 0.835
## # … with 31 more rows

86.1.3 CARE-PF

CARE_medPM <- CARE %>% group_by(ID) %>% summarise(Mean=mean(PM, na.rm=T), Max=max(PM, na.rm=T), Min=min(PM, na.rm=T), Median=median(PM, na.rm=T), StdDev=sd(PM, na.rm=T), Q1=quantile(PM, 0.25, na.rm=T), Q3=quantile(PM, 0.75, na.rm=T))
CARE_medPM %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1  6.37  43.8  1.28   6.17  0.442  5.04  7.94

CARE_medPM <- inner_join(CARE_medPM, CARE, by="ID")
CARE_medPM %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 8 × 8
##   site   Mean   Max   Min Median StdDev    Q1    Q3
##   <fct> <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 101    7.10  9.32  2.50   7.52 0.152   6.35  8.22
## 2 102    5.27  9.34  2.65   4.96 0.153   4.09  6.65
## 3 103    5.43 43.8   1.28   5.69 0.668   4.61  6.44
## 4 104    5.64 19.0   1.35   5.76 0.334   5.09  6.48
## 5 105    7.78 11.2   2.31   8.12 0.365   6.49  9.62
## 6 106    8.38 11.7   2.77   8.34 0.284   7.70  9.21
## 7 107    7.48 12.2   2.22   7.47 0.278   6.25  8.88
## 8 108    5.52  6.98  3.50   5.57 0.0786  5.08  5.94

CARE_medSO4 <- CARE %>% group_by(ID) %>% summarise(Mean=mean(SO4, na.rm=T), Max=max(SO4, na.rm=T), Min=min(SO4, na.rm=T), Median=median(SO4, na.rm=T), StdDev=sd(SO4, na.rm=T), Q1=quantile(SO4, 0.25, na.rm=T), Q3=quantile(SO4, 0.75, na.rm=T))
CARE_medSO4 %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.820  6.46 0.135  0.611 0.0967 0.414  1.28

CARE_medSO4 <- inner_join(CARE_medSO4, CARE, by="ID")
CARE_medSO4 %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 8 × 8
##   site   Mean   Max   Min Median  StdDev    Q1    Q3
##   <fct> <dbl> <dbl> <dbl>  <dbl>   <dbl> <dbl> <dbl>
## 1 101   1.28  2.92  0.354  1.32  0.111   1.07  1.58 
## 2 102   0.454 1.12  0.218  0.435 0.0243  0.376 0.535
## 3 103   0.514 6.46  0.135  0.499 0.0432  0.368 0.671
## 4 104   0.518 6.32  0.188  0.501 0.103   0.382 0.686
## 5 105   1.21  2.69  0.377  1.15  0.131   0.933 1.64 
## 6 106   1.64  3.95  0.417  1.56  0.193   1.32  1.94 
## 7 107   1.14  2.67  0.342  1.04  0.143   0.903 1.28 
## 8 108   0.553 0.767 0.333  0.570 0.00559 0.486 0.600

CARE_medNO3 <- CARE %>% group_by(ID) %>% summarise(Mean=mean(NO3, na.rm=T), Max=max(NO3, na.rm=T), Min=min(NO3, na.rm=T), Median=median(NO3, na.rm=T), StdDev=sd(NO3, na.rm=T), Q1=quantile(NO3, 0.25, na.rm=T), Q3=quantile(NO3, 0.75, na.rm=T))
CARE_medNO3 %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max    Min Median StdDev    Q1    Q3
##   <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.626  2.14 0.0134  0.548 0.0377 0.345 0.968

CARE_medNO3 <- inner_join(CARE_medNO3, CARE, by="ID")
CARE_medNO3 %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 8 × 8
##   site   Mean   Max    Min Median StdDev    Q1    Q3
##   <fct> <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 101   0.910 1.64  0.173   1.02  0.0203 0.809 1.08 
## 2 102   0.588 1.97  0.0798  0.493 0.0426 0.327 0.890
## 3 103   0.368 1.45  0.0134  0.383 0.0282 0.228 0.506
## 4 104   0.395 0.909 0.0393  0.419 0.0297 0.293 0.519
## 5 105   0.795 1.68  0.0501  0.834 0.0557 0.613 1.04 
## 6 106   1.15  2.14  0.179   1.16  0.0537 1.06  1.25 
## 7 107   0.760 1.90  0.0788  0.772 0.0600 0.605 0.922
## 8 108   0.646 1.17  0.137   0.652 0.0130 0.524 0.811

CARE_medNH4 <- CARE %>% group_by(ID) %>% summarise(Mean=mean(NH4, na.rm=T), Max=max(NH4, na.rm=T), Min=min(NH4, na.rm=T), Median=median(NH4, na.rm=T), StdDev=sd(NH4, na.rm=T), Q1=quantile(NH4, 0.25, na.rm=T), Q3=quantile(NH4, 0.75, na.rm=T))
CARE_medNH4 %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.293  1.66     0  0.212 0.0534 0.101 0.511

CARE_medNH4 <- inner_join(CARE_medNH4, CARE, by="ID")
CARE_medNH4 %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 8 × 8
##   site   Mean   Max    Min Median  StdDev     Q1    Q3
##   <fct> <dbl> <dbl>  <dbl>  <dbl>   <dbl>  <dbl> <dbl>
## 1 101   0.496 1.20  0.0290  0.531 0.0495  0.376  0.669
## 2 102   0.203 0.812 0.0455  0.179 0.0305  0.128  0.297
## 3 103   0.146 1.65  0       0.114 0.0340  0.0719 0.224
## 4 104   0.143 1.61  0.0104  0.119 0.0431  0.0700 0.200
## 5 105   0.474 1.40  0.0213  0.440 0.0837  0.300  0.733
## 6 106   0.691 1.66  0       0.652 0.0964  0.528  0.900
## 7 107   0.429 1.31  0.0222  0.381 0.0822  0.287  0.502
## 8 108   0.189 0.341 0.0388  0.205 0.00919 0.135  0.253

CARE_medBC <- CARE %>% group_by(ID) %>% summarise(Mean=mean(BC, na.rm=T), Max=max(BC, na.rm=T), Min=min(BC, na.rm=T), Median=median(BC, na.rm=T), StdDev=sd(BC, na.rm=T), Q1=quantile(BC, 0.25, na.rm=T), Q3=quantile(BC, 0.75, na.rm=T))
CARE_medBC %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max    Min Median StdDev    Q1    Q3
##   <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.469  1.05 0.0401  0.472 0.0205 0.310 0.633

CARE_medBC <- inner_join(CARE_medBC, CARE, by="ID")
CARE_medBC %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 8 × 8
##   site   Mean   Max    Min Median  StdDev    Q1    Q3
##   <fct> <dbl> <dbl>  <dbl>  <dbl>   <dbl> <dbl> <dbl>
## 1 101   0.506 0.804 0.164   0.506 0.0163  0.411 0.629
## 2 102   0.391 0.981 0.141   0.317 0.0114  0.221 0.594
## 3 103   0.439 1.05  0.0401  0.376 0.0210  0.242 0.689
## 4 104   0.473 0.918 0.0648  0.416 0.0171  0.286 0.699
## 5 105   0.514 0.982 0.147   0.523 0.0384  0.409 0.652
## 6 106   0.613 0.977 0.215   0.581 0.0273  0.509 0.728
## 7 107   0.496 0.900 0.156   0.477 0.0288  0.390 0.562
## 8 108   0.341 0.517 0.176   0.341 0.00426 0.287 0.389

CARE_medOM <- CARE %>% group_by(ID) %>% summarise(Mean=mean(OM, na.rm=T), Max=max(OM, na.rm=T), Min=min(OM, na.rm=T), Median=median(OM, na.rm=T), StdDev=sd(OM, na.rm=T), Q1=quantile(OM, 0.25, na.rm=T), Q3=quantile(OM, 0.75, na.rm=T))
CARE_medOM %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max   Min Median StdDev    Q1    Q3
##   <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1  2.97  10.8 0.126   2.75  0.179  1.94  4.20

CARE_medOM <- inner_join(CARE_medOM, CARE, by="ID")
CARE_medOM %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 8 × 8
##   site   Mean   Max   Min Median StdDev    Q1    Q3
##   <fct> <dbl> <dbl> <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 101    2.72  4.87 0.799   2.66 0.0987  2.04  3.61
## 2 102    2.64  7.32 1.00    2.07 0.117   1.33  4.06
## 3 103    2.92 10.8  0.126   2.41 0.269   1.50  4.58
## 4 104    3.18  6.91 0.178   2.82 0.223   1.85  4.60
## 5 105    3.66  7.91 0.890   3.70 0.282   2.65  4.92
## 6 106    3.18  5.80 1.15    2.83 0.132   2.38  4.15
## 7 107    3.54  6.50 0.770   3.42 0.267   2.64  4.42
## 8 108    2.25  3.97 1.34    2.24 0.0945  1.82  2.75

CARE_medSS <- CARE %>% group_by(ID) %>% summarise(Mean=mean(SS, na.rm=T), Max=max(SS, na.rm=T), Min=min(SS, na.rm=T), Median=median(SS, na.rm=T), StdDev=sd(SS, na.rm=T), Q1=quantile(SS, 0.25, na.rm=T), Q3=quantile(SS, 0.75, na.rm=T))
CARE_medSS %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max     Min Median StdDev    Q1    Q3
##   <dbl> <dbl>   <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.243  1.48 0.00537  0.248 0.0151 0.186 0.308

CARE_medSS <- inner_join(CARE_medSS, CARE, by="ID")
CARE_medSS %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 8 × 8
##   site    Mean   Max     Min Median  StdDev     Q1     Q3
##   <fct>  <dbl> <dbl>   <dbl>  <dbl>   <dbl>  <dbl>  <dbl>
## 1 101   0.220  0.334 0.0316  0.228  0.00931 0.173  0.275 
## 2 102   0.0688 0.289 0.00537 0.0540 0.00255 0.0231 0.115 
## 3 103   0.317  1.48  0.0161  0.316  0.0286  0.237  0.369 
## 4 104   0.292  1.12  0.0239  0.307  0.0228  0.224  0.353 
## 5 105   0.229  0.314 0.0719  0.237  0.00669 0.207  0.257 
## 6 106   0.247  0.400 0.0658  0.246  0.0113  0.196  0.290 
## 7 107   0.241  1.10  0.0825  0.243  0.0294  0.220  0.262 
## 8 108   0.0596 0.128 0.0165  0.0514 0.00186 0.0377 0.0865

CARE_medSoil <- CARE %>% group_by(ID) %>% summarise(Mean=mean(Soil, na.rm=T), Max=max(Soil, na.rm=T), Min=min(Soil, na.rm=T), Median=median(Soil, na.rm=T), StdDev=sd(Soil, na.rm=T), Q1=quantile(Soil, 0.25, na.rm=T), Q3=quantile(Soil, 0.75, na.rm=T))
CARE_medSoil %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 1 × 7
##    Mean   Max    Min Median StdDev    Q1    Q3
##   <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl>
## 1 0.326  1.15 0.0117  0.313 0.0188 0.189 0.430

CARE_medSoil <- inner_join(CARE_medSoil, CARE, by="ID")
CARE_medSoil %>% group_by(site) %>% summarise(Mean=mean(Mean, na.rm=T), Max=max(Max, na.rm=T), Min=min(Min, na.rm=T), Median=median(Median, na.rm=T), StdDev=sd(StdDev, na.rm=T), Q1=quantile(Q1, 0.25, na.rm=T), Q3=quantile(Q3, 0.75, na.rm=T))

## # A tibble: 8 × 8
##   site   Mean   Max    Min Median  StdDev    Q1    Q3
##   <fct> <dbl> <dbl>  <dbl>  <dbl>   <dbl> <dbl> <dbl>
## 1 101   0.409 0.809 0.0664  0.417 0.0212  0.266 0.591
## 2 102   0.347 0.705 0.114   0.313 0.00860 0.259 0.460
## 3 103   0.242 1.15  0.0117  0.215 0.0284  0.145 0.350
## 4 104   0.250 1.09  0.0248  0.233 0.0301  0.161 0.346
## 5 105   0.331 0.500 0.0532  0.350 0.00999 0.256 0.421
## 6 106   0.512 0.865 0.0837  0.468 0.0263  0.369 0.700
## 7 107   0.334 0.599 0.0749  0.355 0.0167  0.255 0.429
## 8 108   0.384 0.555 0.203   0.386 0.00486 0.333 0.449

86.1.4 Create summarized time-varying exposures excel file

Simm_medPM <- Simm_medPM %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
Simm_medSO4 <- Simm_medSO4 %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
Simm_medNO3 <- Simm_medNO3 %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
Simm_medNH4 <- Simm_medNH4 %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
Simm_medBC <- Simm_medBC %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
Simm_medOM <- Simm_medOM %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
Simm_medSS <- Simm_medSS %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
Simm_medSoil <- Simm_medSoil %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)

PFF_medPM <- PFF_medPM %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
PFF_medSO4 <- PFF_medSO4 %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
PFF_medNO3 <- PFF_medNO3 %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
PFF_medNH4 <- PFF_medNH4 %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
PFF_medBC <- PFF_medBC %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
PFF_medOM <- PFF_medOM %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
PFF_medSS <- PFF_medSS %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
PFF_medSoil <- PFF_medSoil %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)

CARE_medPM <- CARE_medPM %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
CARE_medSO4 <- CARE_medSO4 %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
CARE_medNO3 <- CARE_medNO3 %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
CARE_medNH4 <- CARE_medNH4 %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
CARE_medBC <- CARE_medBC %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
CARE_medOM <- CARE_medOM %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
CARE_medSS <- CARE_medSS %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)
CARE_medSoil <- CARE_medSoil %>% dplyr::select(ID, Mean, Median, Q1, Q3, StdDev)

Simm_medPM <- Simm_medPM %>% mutate(pollutant="PM")
Simm_medSO4 <- Simm_medSO4 %>% mutate(pollutant="SO4")
Simm_medNO3 <- Simm_medNO3 %>% mutate(pollutant="NO3")
Simm_medNH4 <- Simm_medNH4 %>% mutate(pollutant="NH4")
Simm_medBC <- Simm_medBC %>% mutate(pollutant="BC")
Simm_medOM <- Simm_medOM %>% mutate(pollutant="OM")
Simm_medSS <- Simm_medSS %>% mutate(pollutant="SS")
Simm_medSoil <- Simm_medSoil %>% mutate(pollutant="Soil")

PFF_medPM <- PFF_medPM %>% mutate(pollutant="PM")
PFF_medSO4 <- PFF_medSO4 %>% mutate(pollutant="SO4")
PFF_medNO3 <- PFF_medNO3 %>% mutate(pollutant="NO3")
PFF_medNH4 <- PFF_medNH4 %>% mutate(pollutant="NH4")
PFF_medBC <- PFF_medBC %>% mutate(pollutant="BC")
PFF_medOM <- PFF_medOM %>% mutate(pollutant="OM")
PFF_medSS <- PFF_medSS %>% mutate(pollutant="SS")
PFF_medSoil <- PFF_medSoil %>% mutate(pollutant="Soil")

CARE_medPM <- CARE_medPM %>% mutate(pollutant="PM")
CARE_medSO4 <- CARE_medSO4 %>% mutate(pollutant="SO4")
CARE_medNO3 <- CARE_medNO3 %>% mutate(pollutant="NO3")
CARE_medNH4 <- CARE_medNH4 %>% mutate(pollutant="NH4")
CARE_medBC <- CARE_medBC %>% mutate(pollutant="BC")
CARE_medOM <- CARE_medOM %>% mutate(pollutant="OM")
CARE_medSS <- CARE_medSS %>% mutate(pollutant="SS")
CARE_medSoil <- CARE_medSoil %>% mutate(pollutant="Soil")

Simm_medPM <- Simm_medPM %>% mutate(cohort="Simm")
Simm_medSO4 <- Simm_medSO4 %>% mutate(cohort="Simm")
Simm_medNO3 <- Simm_medNO3 %>% mutate(cohort="Simm")
Simm_medNH4 <- Simm_medNH4 %>% mutate(cohort="Simm")
Simm_medBC <- Simm_medBC %>% mutate(cohort="Simm")
Simm_medOM <- Simm_medOM %>% mutate(cohort="Simm")
Simm_medSS <- Simm_medSS %>% mutate(cohort="Simm")
Simm_medSoil <- Simm_medSoil %>% mutate(cohort="Simm")

PFF_medPM <- PFF_medPM %>% mutate(cohort="PFF")
PFF_medSO4 <- PFF_medSO4 %>% mutate(cohort="PFF")
PFF_medNO3 <- PFF_medNO3 %>% mutate(cohort="PFF")
PFF_medNH4 <- PFF_medNH4 %>% mutate(cohort="PFF")
PFF_medBC <- PFF_medBC %>% mutate(cohort="PFF")
PFF_medOM <- PFF_medOM %>% mutate(cohort="PFF")
PFF_medSS <- PFF_medSS %>% mutate(cohort="PFF")
PFF_medSoil <- PFF_medSoil %>% mutate(cohort="PFF")

CARE_medPM <- CARE_medPM %>% mutate(cohort="CARE")
CARE_medSO4 <- CARE_medSO4 %>% mutate(cohort="CARE")
CARE_medNO3 <- CARE_medNO3 %>% mutate(cohort="CARE")
CARE_medNH4 <- CARE_medNH4 %>% mutate(cohort="CARE")
CARE_medBC <- CARE_medBC %>% mutate(cohort="CARE")
CARE_medOM <- CARE_medOM %>% mutate(cohort="CARE")
CARE_medSS <- CARE_medSS %>% mutate(cohort="CARE")
CARE_medSoil <- CARE_medSoil %>% mutate(cohort="CARE")

PFF_medPM <- PFF_medPM %>% group_by(ID) %>% slice(1)
PFF_medSO4 <- PFF_medSO4 %>% group_by(ID) %>% slice(1)
PFF_medNO3 <- PFF_medNO3 %>% group_by(ID) %>% slice(1)
PFF_medNH4 <- PFF_medNH4 %>% group_by(ID) %>% slice(1)
PFF_medBC <- PFF_medBC %>% group_by(ID) %>% slice(1)
PFF_medOM <- PFF_medOM %>% group_by(ID) %>% slice(1)
PFF_medSS <- PFF_medSS %>% group_by(ID) %>% slice(1)
PFF_medSoil <- PFF_medSoil %>% group_by(ID) %>% slice(1)

CARE_medPM <- CARE_medPM %>% group_by(ID) %>% slice(1)
CARE_medSO4 <- CARE_medSO4 %>% group_by(ID) %>% slice(1)
CARE_medNO3 <- CARE_medNO3 %>% group_by(ID) %>% slice(1)
CARE_medNH4 <- CARE_medNH4 %>% group_by(ID) %>% slice(1)
CARE_medBC <- CARE_medBC %>% group_by(ID) %>% slice(1)
CARE_medOM <- CARE_medOM %>% group_by(ID) %>% slice(1)
CARE_medSS <- CARE_medSS %>% group_by(ID) %>% slice(1)
CARE_medSoil <- CARE_medSoil %>% group_by(ID) %>% slice(1)

Simm_summary <- rbind(Simm_medPM, Simm_medSO4, Simm_medNO3, Simm_medNH4, Simm_medBC, Simm_medOM, Simm_medSS, Simm_medSoil)
PFF_summary <- rbind(PFF_medPM, PFF_medSO4, PFF_medNO3, PFF_medNH4, PFF_medBC, PFF_medOM, PFF_medSS, PFF_medSoil)
CARE_summary <- rbind(CARE_medPM, CARE_medSO4, CARE_medNO3, CARE_medNH4, CARE_medBC, CARE_medOM, CARE_medSS, CARE_medSoil)
CombinedCohorts_summary <- rbind(Simm_summary, PFF_summary, CARE_summary)

write_xlsx(CombinedCohorts_summary, "CombinedCohorts_TimeVaryingSummarizedExposures_2023_03_16.xlsx")

86.2 Summarize Demographics Between Low vs High Exposures

86.2.1 Split Each Cohort into Low vs High PM2.5

low_Simm <- Simm_medPM %>% filter(Median<8)
low_Simm <- left_join(low_Simm, Simm, by="ID")
low_Simm <- low_Simm %>% group_by(ID) %>% slice(1)

high_Simm <- Simm_medPM %>% filter(Median>=8)
high_Simm <- left_join(high_Simm, Simm, by="ID")
high_Simm <- high_Simm %>% group_by(ID) %>% slice(1)

low_PFF <- PFF_medPM %>% filter(Median<8)
low_PFF <- left_join(low_PFF, PFF, by="ID")
low_PFF <- low_PFF %>% group_by(ID) %>% slice(1)

high_PFF <- PFF_medPM %>% filter(Median>=8)
high_PFF <- left_join(high_PFF, PFF, by="ID")
high_PFF <- high_PFF %>% group_by(ID) %>% slice(1)

low_CARE <- CARE_medPM %>% filter(Median<8)
low_CARE <- left_join(low_CARE, CARE, by="ID")
low_CARE <- low_CARE %>% group_by(ID) %>% slice(1)

high_CARE <- CARE_medPM %>% filter(Median>=8)
high_CARE <- left_join(high_CARE, CARE, by="ID")
high_CARE <- high_CARE %>% group_by(ID) %>% slice(1)

86.2.2 Summarizing Function

n_prop_tbl <- function(x) {
  tbl <- table(x)
  res <- cbind(tbl, round(prop.table(tbl)*100,2))
  colnames(res) <-  c('Count', 'Percentage')
  res
}

86.2.3 Sex Breakdown

n_prop_tbl(low_Simm$sex)

##   Count Percentage
## M    90      49.72
## F    91      50.28

n_prop_tbl(high_Simm$sex)

##   Count Percentage
## M   705      56.72
## F   538      43.28

n_prop_tbl(low_PFF$sex.x)

## Warning: Unknown or uninitialised column: `sex.x`.

##      Count Percentage

n_prop_tbl(high_PFF$sex.x)

## Warning: Unknown or uninitialised column: `sex.x`.

##      Count Percentage

n_prop_tbl(low_CARE$sex.x)

## Warning: Unknown or uninitialised column: `sex.x`.

##      Count Percentage

n_prop_tbl(high_CARE$sex.x)

## Warning: Unknown or uninitialised column: `sex.x`.

##      Count Percentage

86.2.4 Race Breakdown

n_prop_tbl(low_Simm$dich_Race)

##           Count Percentage
## White       168      92.82
## Non-White    13       7.18

n_prop_tbl(high_Simm$dich_Race)

##           Count Percentage
## White      1090      87.69
## Non-White   153      12.31

n_prop_tbl(low_PFF$dich_Race.x)

## Warning: Unknown or uninitialised column: `dich_Race.x`.

##      Count Percentage

n_prop_tbl(high_PFF$dich_Race.x)

## Warning: Unknown or uninitialised column: `dich_Race.x`.

##      Count Percentage

n_prop_tbl(low_CARE$dich_Race.x)

## Warning: Unknown or uninitialised column: `dich_Race.x`.

##      Count Percentage

n_prop_tbl(high_CARE$dich_Race.x)

## Warning: Unknown or uninitialised column: `dich_Race.x`.

##      Count Percentage

86.2.5 Smoking History Breakdown

n_prop_tbl(low_Simm$smokeHx)

##         Count Percentage
## Never      69      38.12
## Former     61      33.70
## Always      5       2.76
## Unknown    46      25.41

n_prop_tbl(high_Simm$smokeHx)

##         Count Percentage
## Never     344      27.67
## Former    603      48.51
## Always     33       2.65
## Unknown   263      21.16

n_prop_tbl(low_PFF$smokeHx.x)

## Warning: Unknown or uninitialised column: `smokeHx.x`.

##      Count Percentage

n_prop_tbl(high_PFF$smokeHx.x)

## Warning: Unknown or uninitialised column: `smokeHx.x`.

##      Count Percentage

n_prop_tbl(low_CARE$smokeHx.x)

## Warning: Unknown or uninitialised column: `smokeHx.x`.

##      Count Percentage

n_prop_tbl(high_CARE$smokeHx.x)

## Warning: Unknown or uninitialised column: `smokeHx.x`.

##      Count Percentage

86.2.6 Diagnosis Group Breakdown

n_prop_tbl(low_Simm$dx_group)

##                Count Percentage
## IPF               67      37.02
## CTD-ILD           32      17.68
## HP                14       7.73
## OTHER_IIP          7       3.87
## OTHER_ILD          9       4.97
## PNEUMOCONIOSIS     3       1.66
## UNCLASSIFIABLE    49      27.07

n_prop_tbl(high_Simm$dx_group)

##                Count Percentage
## IPF              649      52.21
## CTD-ILD          268      21.56
## HP                41       3.30
## OTHER_IIP         61       4.91
## OTHER_ILD         41       3.30
## PNEUMOCONIOSIS    23       1.85
## UNCLASSIFIABLE   160      12.87

n_prop_tbl(low_PFF$dx_group.x)

## Warning: Unknown or uninitialised column: `dx_group.x`.

##      Count Percentage

n_prop_tbl(high_PFF$dx_group.x)

## Warning: Unknown or uninitialised column: `dx_group.x`.

##      Count Percentage

n_prop_tbl(low_CARE$dx_group.x)

## Warning: Unknown or uninitialised column: `dx_group.x`.

##      Count Percentage

n_prop_tbl(high_CARE$dx_group.x)

## Warning: Unknown or uninitialised column: `dx_group.x`.

##      Count Percentage

86.3 Age at Enrollment Breakdown

summary(low_Simm$age_dx)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   17.81   59.59   67.30   64.95   72.57   89.23

summary(high_Simm$age_dx)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##   20.22   57.63   65.51   64.10   72.58   94.06       1

summary(low_PFF$age_dx.x)

## Warning: Unknown or uninitialised column: `age_dx.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(high_PFF$age_dx.x)

## Warning: Unknown or uninitialised column: `age_dx.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(low_CARE$age_dx.x)

## Warning: Unknown or uninitialised column: `age_dx.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(high_CARE$age_dx.x)

## Warning: Unknown or uninitialised column: `age_dx.x`.

## Length  Class   Mode 
##      0   NULL   NULL

86.4 Baseline FVC Breakdown

summary(low_Simm$fvc_pct)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##   26.00   58.50   72.00   72.42   86.00  138.00      38

summary(high_Simm$fvc_pct)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##   16.00   53.00   65.00   66.32   79.20  124.17     309

summary(low_PFF$fvc_pct.x)

## Warning: Unknown or uninitialised column: `fvc_pct.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(high_PFF$fvc_pct.x)

## Warning: Unknown or uninitialised column: `fvc_pct.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(low_CARE$fvc_pct.x)

## Warning: Unknown or uninitialised column: `fvc_pct.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(high_CARE$fvc_pct.x)

## Warning: Unknown or uninitialised column: `fvc_pct.x`.

## Length  Class   Mode 
##      0   NULL   NULL

86.5 Baseline DLCO Breakdown

summary(low_Simm$dlco_pct)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##    9.00   38.00   52.50   51.91   63.00  101.00      45

summary(high_Simm$dlco_pct)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##   14.00   36.70   49.00   50.94   63.24  124.00     377

summary(low_PFF$dlco_pct.x)

## Warning: Unknown or uninitialised column: `dlco_pct.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(high_PFF$dlco_pct.x)

## Warning: Unknown or uninitialised column: `dlco_pct.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(low_CARE$dlco_pct.x)

## Warning: Unknown or uninitialised column: `dlco_pct.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(high_CARE$dlco_pct.x)

## Warning: Unknown or uninitialised column: `dlco_pct.x`.

## Length  Class   Mode 
##      0   NULL   NULL

86.6 ECDF-Adjusted Disadvantage Score Breakdown

summary(low_Simm$disadv)

##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max.     NA's 
## 0.007971 0.378986 0.640580 0.581453 0.781884 1.000000        6

summary(high_Simm$disadv)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
## 0.00145 0.24638 0.49928 0.49576 0.74275 1.00000      38

summary(low_PFF$disadv.x)

## Warning: Unknown or uninitialised column: `disadv.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(high_PFF$disadv.x)

## Warning: Unknown or uninitialised column: `disadv.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(low_CARE$disadv.x)

## Warning: Unknown or uninitialised column: `disadv.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(high_CARE$disadv.x)

## Warning: Unknown or uninitialised column: `disadv.x`.

## Length  Class   Mode 
##      0   NULL   NULL

86.7 Time to Censoring Breakdown

summary(low_Simm$time_DeathTxCensor)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##   0.000   1.129   2.789   3.457   4.777  16.225       1

summary(high_Simm$time_DeathTxCensor)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##   0.000   1.213   3.103   4.440   6.446  20.427       7

summary(low_PFF$time_DeathTxCensor.x)

## Warning: Unknown or uninitialised column: `time_DeathTxCensor.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(high_PFF$time_DeathTxCensor.x)

## Warning: Unknown or uninitialised column: `time_DeathTxCensor.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(low_CARE$time_DeathTxCensor.x)

## Warning: Unknown or uninitialised column: `time_DeathTxCensor.x`.

## Length  Class   Mode 
##      0   NULL   NULL

summary(high_CARE$time_DeathTxCensor.x)

## Warning: Unknown or uninitialised column: `time_DeathTxCensor.x`.

## Length  Class   Mode 
##      0   NULL   NULL

86.7.1 Survival Status Breakdown

n_prop_tbl(low_Simm$status)

##   Count Percentage
## 0   135      74.59
## 1    33      18.23
## 2    13       7.18

n_prop_tbl(high_Simm$status)

##   Count Percentage
## 0   381      30.65
## 1   674      54.22
## 2   188      15.12

n_prop_tbl(low_PFF$status.x)

## Warning: Unknown or uninitialised column: `status.x`.

##      Count Percentage

n_prop_tbl(high_PFF$status.x)

## Warning: Unknown or uninitialised column: `status.x`.

##      Count Percentage

n_prop_tbl(low_CARE$status.x)

## Warning: Unknown or uninitialised column: `status.x`.

##      Count Percentage

n_prop_tbl(high_CARE$status.x)

## Warning: Unknown or uninitialised column: `status.x`.

##      Count Percentage

87 Remove unnecessary dataframes

rm(Simm_medPM, Simm_medSO4, Simm_medNO3, Simm_medNH4, Simm_medBC, Simm_medOM, Simm_medSS, Simm_medSoil, PFF_medPM, PFF_medSO4, PFF_medNO3, PFF_medNH4, PFF_medBC, PFF_medOM, PFF_medSS, PFF_medSoil, CARE_medPM, CARE_medSO4, CARE_medNO3, CARE_medNH4, CARE_medBC, CARE_medOM, CARE_medSS, CARE_medSoil)

88 Combined One-Stage Meta-Analysis with Time-Weighted Exposure Estimates

88.1 PM2.5

88.1.1 PM2.5 Continuous Models

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ PM + dx_yr + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM + dx_yr + site, 
##     data = All, id = ID, cluster = cohort)
## 
##   n= 335367, number of events= 6459 
##    (1981 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se        z Pr(>|z|)    
## PM       0.084880  1.088586  0.005861  0.019561    4.339 1.43e-05 ***
## dx_yr    0.238088  1.268821  0.005528  0.137872    1.727  0.08419 .  
## site02R  0.245409  1.278144  0.254458  0.097670    2.513  0.01198 *  
## site03R -0.216998  0.804932  0.263211  0.080355   -2.700  0.00692 ** 
## site04R -0.490774  0.612152  0.271430  0.044364  -11.062  < 2e-16 ***
## site05R -0.535363  0.585457  0.276347  0.112053   -4.778 1.77e-06 ***
## site06R -0.132095  0.876258  0.267152  0.163661   -0.807  0.41959    
## site07R -0.642670  0.525887  0.245835  0.123852   -5.189 2.11e-07 ***
## site09R -0.195108  0.822745  0.269373  0.033389   -5.844 5.11e-09 ***
## site1   -0.380259  0.683684  0.216379  0.001216 -312.734  < 2e-16 ***
## site101 -0.487996  0.613855  0.220322  0.198527   -2.458  0.01397 *  
## site102 -0.420459  0.656745  0.220066  0.201687   -2.085  0.03710 *  
## site103 -0.146977  0.863314  0.215499  0.149057   -0.986  0.32411    
## site104 -0.326472  0.721464  0.221273  0.158771   -2.056  0.03976 *  
## site105 -0.460720  0.630829  0.217575  0.181737   -2.535  0.01124 *  
## site106 -0.572618  0.564047  0.219135  0.137488   -4.165 3.12e-05 ***
## site107 -0.127480  0.880311  0.240844  0.222476   -0.573  0.56664    
## site108 -0.144464  0.865486  0.240161  0.382217   -0.378  0.70546    
## site10R -0.362530  0.695913  0.289827  0.116403   -3.114  0.00184 ** 
## site11R -0.343532  0.709261  0.239099  0.038143   -9.006  < 2e-16 ***
## site12R -0.188462  0.828232  0.252337  0.031612   -5.962 2.49e-09 ***
## site13R -0.581341  0.559148  0.240868  0.116814   -4.977 6.47e-07 ***
## site14R -0.108380  0.897287  0.396068  0.221131   -0.490  0.62405    
## site15R -0.169208  0.844333  0.270954  0.027394   -6.177 6.54e-10 ***
## site16R -0.041370  0.959474  0.263507  0.113692   -0.364  0.71595    
## site17R -0.332971  0.716791  0.275862  0.063096   -5.277 1.31e-07 ***
## site18R -0.670048  0.511684  0.259886  0.208191   -3.218  0.00129 ** 
## site19R -0.041885  0.958980  0.282769  0.065682   -0.638  0.52367    
## site20R -0.380513  0.683511  0.287442  0.044890   -8.476  < 2e-16 ***
## site21R -0.587648  0.555632  0.253164  0.064247   -9.147  < 2e-16 ***
## site22R -0.294333  0.745028  0.247772  0.037573   -7.834 4.74e-15 ***
## site23R -0.360936  0.697024  0.248582  0.034146  -10.570  < 2e-16 ***
## site24R -0.090378  0.913586  0.248862  0.115648   -0.781  0.43451    
## site25R -0.388587  0.678015  0.251283  0.061813   -6.287 3.25e-10 ***
## site26R -0.565617  0.568009  0.265872  0.077287   -7.318 2.51e-13 ***
## site27R -0.352869  0.702669  0.369536  0.033790  -10.443  < 2e-16 ***
## site28R -0.395042  0.673651  0.266491  0.067926   -5.816 6.04e-09 ***
## site29R -0.490035  0.612605  0.335045  0.069483   -7.053 1.76e-12 ***
## site30R -0.639597  0.527505  0.256311  0.082909   -7.714 1.21e-14 ***
## site31R -0.268836  0.764269  0.270671  0.022380  -12.012  < 2e-16 ***
## site32R -0.612004  0.542263  0.268289  0.089629   -6.828 8.60e-12 ***
## site33R -0.456725  0.633354  0.257752  0.043123  -10.591  < 2e-16 ***
## site34R -0.307404  0.735354  0.247754  0.073720   -4.170 3.05e-05 ***
## site35R -0.112425  0.893664  0.264302  0.018541   -6.064 1.33e-09 ***
## site36R -0.482322  0.617348  0.254997  0.059858   -8.058 7.77e-16 ***
## site37R -0.519386  0.594886  0.257370  0.068392   -7.594 3.10e-14 ***
## site38R -0.447477  0.639239  0.261378  0.026047  -17.180  < 2e-16 ***
## site39R -0.358344  0.698833  0.289796  0.061832   -5.795 6.81e-09 ***
## site40R -0.541991  0.581589  0.268845  0.114117   -4.749 2.04e-06 ***
## site41R -0.466901  0.626942  0.252030  0.043890  -10.638  < 2e-16 ***
## site42R -0.135764  0.873049  0.275877  0.093730   -1.448  0.14749    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## PM         1.0886     0.9186    1.0476    1.1311
## dx_yr      1.2688     0.7881    0.9684    1.6625
## site02R    1.2781     0.7824    1.0555    1.5478
## site03R    0.8049     1.2423    0.6876    0.9422
## site04R    0.6122     1.6336    0.5612    0.6678
## site05R    0.5855     1.7081    0.4700    0.7292
## site06R    0.8763     1.1412    0.6358    1.2076
## site07R    0.5259     1.9016    0.4125    0.6704
## site09R    0.8227     1.2154    0.7706    0.8784
## site1      0.6837     1.4627    0.6821    0.6853
## site101    0.6139     1.6290    0.4160    0.9058
## site102    0.6567     1.5227    0.4423    0.9752
## site103    0.8633     1.1583    0.6446    1.1562
## site104    0.7215     1.3861    0.5285    0.9848
## site105    0.6308     1.5852    0.4418    0.9008
## site106    0.5640     1.7729    0.4308    0.7385
## site107    0.8803     1.1360    0.5692    1.3615
## site108    0.8655     1.1554    0.4092    1.8306
## site10R    0.6959     1.4370    0.5540    0.8743
## site11R    0.7093     1.4099    0.6582    0.7643
## site12R    0.8282     1.2074    0.7785    0.8812
## site13R    0.5591     1.7884    0.4447    0.7030
## site14R    0.8973     1.1145    0.5817    1.3841
## site15R    0.8443     1.1844    0.8002    0.8909
## site16R    0.9595     1.0422    0.7678    1.1990
## site17R    0.7168     1.3951    0.6334    0.8111
## site18R    0.5117     1.9543    0.3402    0.7695
## site19R    0.9590     1.0428    0.8431    1.0907
## site20R    0.6835     1.4630    0.6259    0.7464
## site21R    0.5556     1.7998    0.4899    0.6302
## site22R    0.7450     1.3422    0.6921    0.8020
## site23R    0.6970     1.4347    0.6519    0.7453
## site24R    0.9136     1.0946    0.7283    1.1460
## site25R    0.6780     1.4749    0.6007    0.7653
## site26R    0.5680     1.7605    0.4882    0.6609
## site27R    0.7027     1.4231    0.6576    0.7508
## site28R    0.6737     1.4844    0.5897    0.7696
## site29R    0.6126     1.6324    0.5346    0.7020
## site30R    0.5275     1.8957    0.4484    0.6206
## site31R    0.7643     1.3084    0.7315    0.7985
## site32R    0.5423     1.8441    0.4549    0.6464
## site33R    0.6334     1.5789    0.5820    0.6892
## site34R    0.7354     1.3599    0.6364    0.8497
## site35R    0.8937     1.1190    0.8618    0.9267
## site36R    0.6173     1.6198    0.5490    0.6942
## site37R    0.5949     1.6810    0.5203    0.6802
## site38R    0.6392     1.5644    0.6074    0.6727
## site39R    0.6988     1.4310    0.6191    0.7889
## site40R    0.5816     1.7194    0.4650    0.7274
## site41R    0.6269     1.5950    0.5753    0.6833
## site42R    0.8730     1.1454    0.7265    1.0491
## 
## Concordance= 0.672  (se = 0.065 )
## Likelihood ratio test= 2514  on 51 df,   p=<2e-16
## Wald test            = 135.9  on 51 df,   p=1e-09
## Score (logrank) test = 2141  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ PM + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = All, id = ID, 
##     cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## PM                  0.088556  1.092596  0.006029  0.020364   4.349 1.37e-05 ***
## dx_yr               0.235450  1.265478  0.005680  0.140489   1.676 0.093752 .  
## age_dx              0.010062  1.010113  0.001176  0.003891   2.586 0.009706 ** 
## sexF               -0.142647  0.867060  0.026705  0.059993  -2.378 0.017420 *  
## dich_RaceNon-White -0.019531  0.980659  0.036895  0.024148  -0.809 0.418635    
## smokeHxFormer       0.099200  1.104287  0.032889  0.048676   2.038 0.041553 *  
## smokeHxAlways       0.031820  1.032332  0.074835  0.037264   0.854 0.393158    
## smokeHxUnknown      0.032399  1.032930  0.070636  0.045101   0.718 0.472523    
## smokeHxEver        -0.010767  0.989290  0.049938  0.014588  -0.738 0.460464    
## disadv             -0.005585  0.994431  0.045354  0.056866  -0.098 0.921764    
## site02R             0.211792  1.235891  0.258961  0.091144   2.324 0.020141 *  
## site03R            -0.207747  0.812413  0.267561  0.078655  -2.641 0.008260 ** 
## site04R            -0.463205  0.629263  0.275735  0.020666 -22.414  < 2e-16 ***
## site05R            -0.522535  0.593015  0.282492  0.134040  -3.898 9.69e-05 ***
## site06R            -0.177896  0.837030  0.273135  0.200089  -0.889 0.373959    
## site07R            -0.696435  0.498359  0.250491  0.112083  -6.214 5.18e-10 ***
## site09R            -0.252072  0.777189  0.275700  0.026438  -9.535  < 2e-16 ***
## site1              -0.451332  0.636779  0.223822  0.024934 -18.101  < 2e-16 ***
## site101            -0.509880  0.600568  0.227123  0.171241  -2.978 0.002906 ** 
## site102            -0.460996  0.630655  0.227214  0.182803  -2.522 0.011675 *  
## site103            -0.174666  0.839737  0.222667  0.116934  -1.494 0.135249    
## site104            -0.381590  0.682775  0.228202  0.120817  -3.158 0.001586 ** 
## site105            -0.520288  0.594350  0.224993  0.148989  -3.492 0.000479 ***
## site106            -0.686344  0.503413  0.226294  0.096139  -7.139 9.40e-13 ***
## site107            -0.189419  0.827440  0.247433  0.192632  -0.983 0.325451    
## site108            -0.212134  0.808857  0.246457  0.348456  -0.609 0.542669    
## site10R            -0.422988  0.655087  0.296657  0.104715  -4.039 5.36e-05 ***
## site11R            -0.423571  0.654705  0.245575  0.039387 -10.754  < 2e-16 ***
## site12R            -0.239357  0.787134  0.256966  0.044150  -5.421 5.91e-08 ***
## site13R            -0.674803  0.509257  0.245602  0.110496  -6.107 1.02e-09 ***
## site14R            -0.105293  0.900061  0.398980  0.222689  -0.473 0.636339    
## site15R            -0.245542  0.782281  0.275615  0.043929  -5.590 2.28e-08 ***
## site16R            -0.028750  0.971660  0.267664  0.113909  -0.252 0.800738    
## site17R            -0.306669  0.735894  0.280211  0.059062  -5.192 2.08e-07 ***
## site18R            -0.722091  0.485736  0.264575  0.187169  -3.858 0.000114 ***
## site19R            -0.093466  0.910769  0.287043  0.046770  -1.998 0.045671 *  
## site20R            -0.411981  0.662337  0.291694  0.046621  -8.837  < 2e-16 ***
## site21R            -0.642845  0.525795  0.257732  0.082188  -7.822 5.21e-15 ***
## site22R            -0.363549  0.695205  0.252567  0.048145  -7.551 4.31e-14 ***
## site23R            -0.394406  0.674080  0.255427  0.052187  -7.558 4.11e-14 ***
## site24R            -0.072320  0.930234  0.253705  0.117750  -0.614 0.539096    
## site25R            -0.437696  0.645522  0.257596  0.066434  -6.588 4.45e-11 ***
## site26R            -0.626526  0.534445  0.270456  0.071865  -8.718  < 2e-16 ***
## site27R            -0.347629  0.706361  0.372620  0.020876 -16.652  < 2e-16 ***
## site28R            -0.332687  0.716994  0.271033  0.061239  -5.433 5.55e-08 ***
## site29R            -0.540347  0.582546  0.339165  0.074672  -7.236 4.61e-13 ***
## site30R            -0.708897  0.492187  0.261520  0.080032  -8.858  < 2e-16 ***
## site31R            -0.272634  0.761371  0.274821  0.014096 -19.341  < 2e-16 ***
## site32R            -0.620950  0.537434  0.278278  0.065385  -9.497  < 2e-16 ***
## site33R            -0.502324  0.605123  0.263267  0.035671 -14.082  < 2e-16 ***
## site34R            -0.360238  0.697510  0.252924  0.066956  -5.380 7.44e-08 ***
## site35R            -0.175949  0.838661  0.270129  0.015505 -11.348  < 2e-16 ***
## site36R            -0.496404  0.608716  0.259614  0.067086  -7.400 1.37e-13 ***
## site37R            -0.578580  0.560694  0.261900  0.065259  -8.866  < 2e-16 ***
## site38R            -0.480567  0.618433  0.266917  0.022155 -21.691  < 2e-16 ***
## site39R            -0.424512  0.654089  0.294033  0.043745  -9.704  < 2e-16 ***
## site40R            -0.620467  0.537693  0.274443  0.110308  -5.625 1.86e-08 ***
## site41R            -0.487068  0.614425  0.256422  0.037283 -13.064  < 2e-16 ***
## site42R            -0.149197  0.861399  0.279960  0.096399  -1.548 0.121695    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## PM                    1.0926     0.9153    1.0498    1.1371
## dx_yr                 1.2655     0.7902    0.9609    1.6666
## age_dx                1.0101     0.9900    1.0024    1.0178
## sexF                  0.8671     1.1533    0.7709    0.9752
## dich_RaceNon-White    0.9807     1.0197    0.9353    1.0282
## smokeHxFormer         1.1043     0.9056    1.0038    1.2148
## smokeHxAlways         1.0323     0.9687    0.9596    1.1106
## smokeHxUnknown        1.0329     0.9681    0.9455    1.1284
## smokeHxEver           0.9893     1.0108    0.9614    1.0180
## disadv                0.9944     1.0056    0.8895    1.1117
## site02R               1.2359     0.8091    1.0337    1.4776
## site03R               0.8124     1.2309    0.6963    0.9478
## site04R               0.6293     1.5892    0.6043    0.6553
## site05R               0.5930     1.6863    0.4560    0.7712
## site06R               0.8370     1.1947    0.5655    1.2390
## site07R               0.4984     2.0066    0.4001    0.6208
## site09R               0.7772     1.2867    0.7379    0.8185
## site1                 0.6368     1.5704    0.6064    0.6687
## site101               0.6006     1.6651    0.4293    0.8401
## site102               0.6307     1.5857    0.4407    0.9024
## site103               0.8397     1.1908    0.6677    1.0560
## site104               0.6828     1.4646    0.5388    0.8652
## site105               0.5943     1.6825    0.4438    0.7959
## site106               0.5034     1.9864    0.4170    0.6078
## site107               0.8274     1.2085    0.5672    1.2070
## site108               0.8089     1.2363    0.4086    1.6013
## site10R               0.6551     1.5265    0.5335    0.8043
## site11R               0.6547     1.5274    0.6061    0.7072
## site12R               0.7871     1.2704    0.7219    0.8583
## site13R               0.5093     1.9636    0.4101    0.6324
## site14R               0.9001     1.1110    0.5817    1.3926
## site15R               0.7823     1.2783    0.7177    0.8526
## site16R               0.9717     1.0292    0.7772    1.2147
## site17R               0.7359     1.3589    0.6555    0.8262
## site18R               0.4857     2.0587    0.3366    0.7010
## site19R               0.9108     1.0980    0.8310    0.9982
## site20R               0.6623     1.5098    0.6045    0.7257
## site21R               0.5258     1.9019    0.4476    0.6177
## site22R               0.6952     1.4384    0.6326    0.7640
## site23R               0.6741     1.4835    0.6085    0.7467
## site24R               0.9302     1.0750    0.7385    1.1717
## site25R               0.6455     1.5491    0.5667    0.7353
## site26R               0.5344     1.8711    0.4642    0.6153
## site27R               0.7064     1.4157    0.6780    0.7359
## site28R               0.7170     1.3947    0.6359    0.8084
## site29R               0.5825     1.7166    0.5032    0.6744
## site30R               0.4922     2.0317    0.4207    0.5758
## site31R               0.7614     1.3134    0.7406    0.7827
## site32R               0.5374     1.8607    0.4728    0.6109
## site33R               0.6051     1.6526    0.5643    0.6489
## site34R               0.6975     1.4337    0.6117    0.7953
## site35R               0.8387     1.1924    0.8136    0.8645
## site36R               0.6087     1.6428    0.5337    0.6943
## site37R               0.5607     1.7835    0.4934    0.6372
## site38R               0.6184     1.6170    0.5922    0.6459
## site39R               0.6541     1.5288    0.6003    0.7126
## site40R               0.5377     1.8598    0.4332    0.6675
## site41R               0.6144     1.6275    0.5711    0.6610
## site42R               0.8614     1.1609    0.7131    1.0405
## 
## Concordance= 0.676  (se = 0.062 )
## Likelihood ratio test= 2687  on 59 df,   p=<2e-16
## Wald test            = 122.2  on 59 df,   p=3e-06
## Score (logrank) test = 2291  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

88.1.2 PM2.5 Per IQR

summary(All$PM)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##   1.275   6.121   8.003   8.101   9.580  43.788     688

IQR(All$PM, na.rm=T)

## [1] 3.459232

# Will use the 5yr pre-censoring IQR (2.592959), not this one
All <- All %>% mutate(PM_IQR = PM/2.592959)

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ PM_IQR + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM_IQR + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## PM_IQR              0.229623  1.258125  0.015633  0.052803   4.349 1.37e-05 ***
## dx_yr               0.235450  1.265478  0.005680  0.140489   1.676 0.093752 .  
## age_dx              0.010062  1.010113  0.001176  0.003891   2.586 0.009706 ** 
## sexF               -0.142647  0.867060  0.026705  0.059993  -2.378 0.017420 *  
## dich_RaceNon-White -0.019531  0.980659  0.036895  0.024148  -0.809 0.418635    
## smokeHxFormer       0.099200  1.104287  0.032889  0.048676   2.038 0.041553 *  
## smokeHxAlways       0.031820  1.032332  0.074835  0.037264   0.854 0.393158    
## smokeHxUnknown      0.032399  1.032930  0.070636  0.045101   0.718 0.472523    
## smokeHxEver        -0.010767  0.989290  0.049938  0.014588  -0.738 0.460464    
## disadv             -0.005585  0.994431  0.045354  0.056866  -0.098 0.921764    
## site02R             0.211792  1.235891  0.258961  0.091144   2.324 0.020141 *  
## site03R            -0.207747  0.812413  0.267561  0.078655  -2.641 0.008260 ** 
## site04R            -0.463205  0.629263  0.275735  0.020666 -22.414  < 2e-16 ***
## site05R            -0.522535  0.593015  0.282492  0.134040  -3.898 9.69e-05 ***
## site06R            -0.177896  0.837030  0.273135  0.200089  -0.889 0.373959    
## site07R            -0.696435  0.498359  0.250491  0.112083  -6.214 5.18e-10 ***
## site09R            -0.252072  0.777189  0.275700  0.026438  -9.535  < 2e-16 ***
## site1              -0.451332  0.636779  0.223822  0.024934 -18.101  < 2e-16 ***
## site101            -0.509880  0.600568  0.227123  0.171241  -2.978 0.002906 ** 
## site102            -0.460996  0.630655  0.227214  0.182803  -2.522 0.011675 *  
## site103            -0.174666  0.839737  0.222667  0.116934  -1.494 0.135249    
## site104            -0.381590  0.682775  0.228202  0.120817  -3.158 0.001586 ** 
## site105            -0.520288  0.594350  0.224993  0.148989  -3.492 0.000479 ***
## site106            -0.686344  0.503413  0.226294  0.096139  -7.139 9.40e-13 ***
## site107            -0.189419  0.827440  0.247433  0.192632  -0.983 0.325451    
## site108            -0.212134  0.808857  0.246457  0.348456  -0.609 0.542669    
## site10R            -0.422988  0.655087  0.296657  0.104715  -4.039 5.36e-05 ***
## site11R            -0.423571  0.654705  0.245575  0.039387 -10.754  < 2e-16 ***
## site12R            -0.239357  0.787134  0.256966  0.044150  -5.421 5.91e-08 ***
## site13R            -0.674803  0.509257  0.245602  0.110496  -6.107 1.02e-09 ***
## site14R            -0.105293  0.900061  0.398980  0.222689  -0.473 0.636339    
## site15R            -0.245542  0.782281  0.275615  0.043929  -5.590 2.28e-08 ***
## site16R            -0.028750  0.971660  0.267664  0.113909  -0.252 0.800738    
## site17R            -0.306669  0.735894  0.280211  0.059062  -5.192 2.08e-07 ***
## site18R            -0.722091  0.485736  0.264575  0.187169  -3.858 0.000114 ***
## site19R            -0.093466  0.910769  0.287043  0.046770  -1.998 0.045671 *  
## site20R            -0.411981  0.662337  0.291694  0.046621  -8.837  < 2e-16 ***
## site21R            -0.642845  0.525795  0.257732  0.082188  -7.822 5.21e-15 ***
## site22R            -0.363549  0.695205  0.252567  0.048145  -7.551 4.31e-14 ***
## site23R            -0.394406  0.674080  0.255427  0.052187  -7.558 4.11e-14 ***
## site24R            -0.072320  0.930234  0.253705  0.117750  -0.614 0.539096    
## site25R            -0.437696  0.645522  0.257596  0.066434  -6.588 4.45e-11 ***
## site26R            -0.626526  0.534445  0.270456  0.071865  -8.718  < 2e-16 ***
## site27R            -0.347629  0.706361  0.372620  0.020876 -16.652  < 2e-16 ***
## site28R            -0.332687  0.716994  0.271033  0.061239  -5.433 5.55e-08 ***
## site29R            -0.540347  0.582546  0.339165  0.074672  -7.236 4.61e-13 ***
## site30R            -0.708897  0.492187  0.261520  0.080032  -8.858  < 2e-16 ***
## site31R            -0.272634  0.761371  0.274821  0.014096 -19.341  < 2e-16 ***
## site32R            -0.620950  0.537434  0.278278  0.065385  -9.497  < 2e-16 ***
## site33R            -0.502324  0.605123  0.263267  0.035671 -14.082  < 2e-16 ***
## site34R            -0.360238  0.697510  0.252924  0.066956  -5.380 7.44e-08 ***
## site35R            -0.175949  0.838661  0.270129  0.015505 -11.348  < 2e-16 ***
## site36R            -0.496404  0.608716  0.259614  0.067086  -7.400 1.37e-13 ***
## site37R            -0.578580  0.560694  0.261900  0.065259  -8.866  < 2e-16 ***
## site38R            -0.480567  0.618433  0.266917  0.022155 -21.691  < 2e-16 ***
## site39R            -0.424512  0.654089  0.294033  0.043745  -9.704  < 2e-16 ***
## site40R            -0.620467  0.537693  0.274443  0.110308  -5.625 1.86e-08 ***
## site41R            -0.487068  0.614425  0.256422  0.037283 -13.064  < 2e-16 ***
## site42R            -0.149197  0.861399  0.279960  0.096399  -1.548 0.121695    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## PM_IQR                1.2581     0.7948    1.1344    1.3953
## dx_yr                 1.2655     0.7902    0.9609    1.6666
## age_dx                1.0101     0.9900    1.0024    1.0178
## sexF                  0.8671     1.1533    0.7709    0.9752
## dich_RaceNon-White    0.9807     1.0197    0.9353    1.0282
## smokeHxFormer         1.1043     0.9056    1.0038    1.2148
## smokeHxAlways         1.0323     0.9687    0.9596    1.1106
## smokeHxUnknown        1.0329     0.9681    0.9455    1.1284
## smokeHxEver           0.9893     1.0108    0.9614    1.0180
## disadv                0.9944     1.0056    0.8895    1.1117
## site02R               1.2359     0.8091    1.0337    1.4776
## site03R               0.8124     1.2309    0.6963    0.9478
## site04R               0.6293     1.5892    0.6043    0.6553
## site05R               0.5930     1.6863    0.4560    0.7712
## site06R               0.8370     1.1947    0.5655    1.2390
## site07R               0.4984     2.0066    0.4001    0.6208
## site09R               0.7772     1.2867    0.7379    0.8185
## site1                 0.6368     1.5704    0.6064    0.6687
## site101               0.6006     1.6651    0.4293    0.8401
## site102               0.6307     1.5857    0.4407    0.9024
## site103               0.8397     1.1908    0.6677    1.0560
## site104               0.6828     1.4646    0.5388    0.8652
## site105               0.5943     1.6825    0.4438    0.7959
## site106               0.5034     1.9864    0.4170    0.6078
## site107               0.8274     1.2085    0.5672    1.2070
## site108               0.8089     1.2363    0.4086    1.6013
## site10R               0.6551     1.5265    0.5335    0.8043
## site11R               0.6547     1.5274    0.6061    0.7072
## site12R               0.7871     1.2704    0.7219    0.8583
## site13R               0.5093     1.9636    0.4101    0.6324
## site14R               0.9001     1.1110    0.5817    1.3926
## site15R               0.7823     1.2783    0.7177    0.8526
## site16R               0.9717     1.0292    0.7772    1.2147
## site17R               0.7359     1.3589    0.6555    0.8262
## site18R               0.4857     2.0587    0.3366    0.7010
## site19R               0.9108     1.0980    0.8310    0.9982
## site20R               0.6623     1.5098    0.6045    0.7257
## site21R               0.5258     1.9019    0.4476    0.6177
## site22R               0.6952     1.4384    0.6326    0.7640
## site23R               0.6741     1.4835    0.6085    0.7467
## site24R               0.9302     1.0750    0.7385    1.1717
## site25R               0.6455     1.5491    0.5667    0.7353
## site26R               0.5344     1.8711    0.4642    0.6153
## site27R               0.7064     1.4157    0.6780    0.7359
## site28R               0.7170     1.3947    0.6359    0.8084
## site29R               0.5825     1.7166    0.5032    0.6744
## site30R               0.4922     2.0317    0.4207    0.5758
## site31R               0.7614     1.3134    0.7406    0.7827
## site32R               0.5374     1.8607    0.4728    0.6109
## site33R               0.6051     1.6526    0.5643    0.6489
## site34R               0.6975     1.4337    0.6117    0.7953
## site35R               0.8387     1.1924    0.8136    0.8645
## site36R               0.6087     1.6428    0.5337    0.6943
## site37R               0.5607     1.7835    0.4934    0.6372
## site38R               0.6184     1.6170    0.5922    0.6459
## site39R               0.6541     1.5288    0.6003    0.7126
## site40R               0.5377     1.8598    0.4332    0.6675
## site41R               0.6144     1.6275    0.5711    0.6610
## site42R               0.8614     1.1609    0.7131    1.0405
## 
## Concordance= 0.676  (se = 0.062 )
## Likelihood ratio test= 2687  on 59 df,   p=<2e-16
## Wald test            = 122.2  on 59 df,   p=3e-06
## Score (logrank) test = 2291  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

So this indicates that there is a HR of 1.26 per IQR increase in PM2.5 as compared with a HR of 1.09 per 1ug/m3 increase in PM2.5.

88.1.3 PM2.5 Variable HR Spline Models

Base model

#First need to make dataframe that only includes patients with a value for event
Allx <- All %>% filter(!is.na(PM) & !is.na(deadORtx) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(cohort) & !is.na(site) & PM<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$PM, df=3) + Allx$dx_yr + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$PM, exp(predicted$fit), type="n", xlim=c(0,20), ylim=c(0,15))
lines(sm.spline(Allx$PM, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$PM, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$PM, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

Complete model

#First need to make dataframe that only includes patients with time_DeathTxCensor
Allx <- All %>% filter(!is.na(PM) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(deadORtx) & !is.na(age_dx) & !is.na(sex) & !is.na(smokeHx) & !is.na(dich_Race) & !is.na(disadv) & !is.na(site) & PM<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$PM, df=3) + Allx$dx_yr + Allx$age_dx + Allx$sex + Allx$smokeHx + Allx$dich_Race + Allx$disadv + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$PM, exp(predicted$fit), type="n", xlim=c(0,20), ylim=c(0,10))
lines(sm.spline(Allx$PM, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$PM, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$PM, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

88.2 SO4

88.2.1 SO4 Continuous Models

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ SO4 + dx_yr + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4 + dx_yr + 
##     site, data = All, id = ID, cluster = cohort)
## 
##   n= 335367, number of events= 6459 
##    (1981 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## SO4      0.748461  2.113744  0.022746  0.089857   8.329  < 2e-16 ***
## dx_yr    0.347537  1.415577  0.006923  0.123523   2.814 0.004900 ** 
## site02R  0.152264  1.164468  0.254111  0.108620   1.402 0.160974    
## site03R -0.439030  0.644662  0.263350  0.121680  -3.608 0.000308 ***
## site04R -1.098973  0.333213  0.271994  0.133146  -8.254  < 2e-16 ***
## site05R -1.483008  0.226954  0.278023  0.096035 -15.442  < 2e-16 ***
## site06R -0.928201  0.395264  0.268446  0.073787 -12.579  < 2e-16 ***
## site07R -1.348089  0.259736  0.246965  0.208336  -6.471 9.75e-11 ***
## site09R -0.700805  0.496186  0.269894  0.082708  -8.473  < 2e-16 ***
## site1   -1.516675  0.219440  0.220067  0.100879 -15.035  < 2e-16 ***
## site101 -1.112890  0.328608  0.221349  0.228563  -4.869 1.12e-06 ***
## site102 -0.599315  0.549188  0.220135  0.158515  -3.781 0.000156 ***
## site103 -0.295001  0.744531  0.215592  0.132142  -2.232 0.025585 *  
## site104 -0.595559  0.551255  0.222107  0.162758  -3.659 0.000253 ***
## site105 -0.964830  0.381048  0.218249  0.214658  -4.495 6.97e-06 ***
## site106 -1.258365  0.284118  0.220345  0.205157  -6.134 8.59e-10 ***
## site107 -0.666938  0.513278  0.241526  0.229282  -2.909 0.003628 ** 
## site108 -0.580101  0.559842  0.240487  0.302802  -1.916 0.055394 .  
## site10R -0.841165  0.431208  0.290331  0.167055  -5.035 4.77e-07 ***
## site11R -1.100279  0.332778  0.240375  0.134139  -8.203 2.35e-16 ***
## site12R -0.588863  0.554958  0.252693  0.049133 -11.985  < 2e-16 ***
## site13R -1.361731  0.256217  0.242270  0.191679  -7.104 1.21e-12 ***
## site14R -0.747290  0.473648  0.396600  0.112661  -6.633 3.29e-11 ***
## site15R -0.838237  0.432472  0.271845  0.098381  -8.520  < 2e-16 ***
## site16R -0.565078  0.568316  0.264023  0.050194 -11.258  < 2e-16 ***
## site17R -1.071869  0.342368  0.276904  0.100849 -10.628  < 2e-16 ***
## site18R -1.423026  0.240984  0.261137  0.277700  -5.124 2.99e-07 ***
## site19R -0.254052  0.775651  0.282815  0.077106  -3.295 0.000985 ***
## site20R -0.980594  0.375088  0.288115  0.083073 -11.804  < 2e-16 ***
## site21R -1.314766  0.268537  0.254172  0.107753 -12.202  < 2e-16 ***
## site22R -0.945716  0.388401  0.248622  0.088630 -10.670  < 2e-16 ***
## site23R -1.067863  0.343742  0.249665  0.104785 -10.191  < 2e-16 ***
## site24R -0.445988  0.640191  0.249109  0.039215 -11.373  < 2e-16 ***
## site25R -0.420806  0.656517  0.250645  0.110280  -3.816 0.000136 ***
## site26R -1.438868  0.237196  0.267596  0.209953  -6.853 7.22e-12 ***
## site27R -1.140559  0.319640  0.370458  0.126148  -9.041  < 2e-16 ***
## site28R -1.351952  0.258735  0.268286  0.122315 -11.053  < 2e-16 ***
## site29R -1.101639  0.332326  0.335670  0.146025  -7.544 4.55e-14 ***
## site30R -1.250839  0.286265  0.256798  0.201281  -6.214 5.15e-10 ***
## site31R -1.036938  0.354539  0.271796  0.083542 -12.412  < 2e-16 ***
## site32R -0.741550  0.476375  0.268181  0.139929  -5.299 1.16e-07 ***
## site33R -1.290355  0.275173  0.259306  0.166805  -7.736 1.03e-14 ***
## site34R -0.982762  0.374276  0.248743  0.140681  -6.986 2.83e-12 ***
## site35R -0.551403  0.576141  0.264656  0.059851  -9.213  < 2e-16 ***
## site36R -1.397454  0.247226  0.256778  0.113584 -12.303  < 2e-16 ***
## site37R -1.337976  0.262376  0.258777  0.170555  -7.845 4.34e-15 ***
## site38R -1.150914  0.316347  0.262414  0.097680 -11.783  < 2e-16 ***
## site39R -0.438723  0.644859  0.289769  0.090481  -4.849 1.24e-06 ***
## site40R -0.681071  0.506075  0.266599  0.204503  -3.330 0.000867 ***
## site41R -1.011483  0.363679  0.252629  0.110912  -9.120  < 2e-16 ***
## site42R -0.915697  0.400237  0.277037  0.067731 -13.520  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## SO4        2.1137     0.4731    1.7724    2.5208
## dx_yr      1.4156     0.7064    1.1112    1.8033
## site02R    1.1645     0.8588    0.9412    1.4407
## site03R    0.6447     1.5512    0.5079    0.8183
## site04R    0.3332     3.0011    0.2567    0.4326
## site05R    0.2270     4.4062    0.1880    0.2740
## site06R    0.3953     2.5300    0.3420    0.4568
## site07R    0.2597     3.8501    0.1727    0.3907
## site09R    0.4962     2.0154    0.4219    0.5835
## site1      0.2194     4.5570    0.1801    0.2674
## site101    0.3286     3.0431    0.2100    0.5143
## site102    0.5492     1.8209    0.4025    0.7493
## site103    0.7445     1.3431    0.5746    0.9646
## site104    0.5513     1.8140    0.4007    0.7584
## site105    0.3810     2.6243    0.2502    0.5804
## site106    0.2841     3.5197    0.1901    0.4247
## site107    0.5133     1.9483    0.3275    0.8045
## site108    0.5598     1.7862    0.3093    1.0135
## site10R    0.4312     2.3191    0.3108    0.5983
## site11R    0.3328     3.0050    0.2558    0.4328
## site12R    0.5550     1.8019    0.5040    0.6111
## site13R    0.2562     3.9029    0.1760    0.3730
## site14R    0.4736     2.1113    0.3798    0.5907
## site15R    0.4325     2.3123    0.3566    0.5244
## site16R    0.5683     1.7596    0.5151    0.6271
## site17R    0.3424     2.9208    0.2810    0.4172
## site18R    0.2410     4.1497    0.1398    0.4153
## site19R    0.7757     1.2892    0.6669    0.9022
## site20R    0.3751     2.6660    0.3187    0.4414
## site21R    0.2685     3.7239    0.2174    0.3317
## site22R    0.3884     2.5747    0.3265    0.4621
## site23R    0.3437     2.9092    0.2799    0.4221
## site24R    0.6402     1.5620    0.5928    0.6913
## site25R    0.6565     1.5232    0.5289    0.8149
## site26R    0.2372     4.2159    0.1572    0.3579
## site27R    0.3196     3.1285    0.2496    0.4093
## site28R    0.2587     3.8650    0.2036    0.3288
## site29R    0.3323     3.0091    0.2496    0.4424
## site30R    0.2863     3.4933    0.1929    0.4247
## site31R    0.3545     2.8206    0.3010    0.4176
## site32R    0.4764     2.0992    0.3621    0.6267
## site33R    0.2752     3.6341    0.1984    0.3816
## site34R    0.3743     2.6718    0.2841    0.4931
## site35R    0.5761     1.7357    0.5124    0.6478
## site36R    0.2472     4.0449    0.1979    0.3089
## site37R    0.2624     3.8113    0.1878    0.3665
## site38R    0.3163     3.1611    0.2612    0.3831
## site39R    0.6449     1.5507    0.5401    0.7700
## site40R    0.5061     1.9760    0.3390    0.7556
## site41R    0.3637     2.7497    0.2926    0.4520
## site42R    0.4002     2.4985    0.3505    0.4571
## 
## Concordance= 0.68  (se = 0.066 )
## Likelihood ratio test= 3031  on 51 df,   p=<2e-16
## Wald test            = 220.3  on 51 df,   p=<2e-16
## Score (logrank) test = 2466  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SO4 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)  robust se       z Pr(>|z|)
## SO4                 0.7486195  2.1140794  0.0228978  0.0801980   9.335  < 2e-16
## dx_yr               0.3436630  1.4101033  0.0070602  0.1252699   2.743 0.006081
## age_dx              0.0091231  1.0091649  0.0011770  0.0034138   2.672 0.007531
## sexF               -0.1358048  0.8730131  0.0267003  0.0629412  -2.158 0.030955
## dich_RaceNon-White  0.0173323  1.0174833  0.0368078  0.0356521   0.486 0.626861
## smokeHxFormer       0.1214572  1.1291410  0.0328996  0.0232463   5.225 1.74e-07
## smokeHxAlways       0.0250676  1.0253845  0.0748671  0.0601250   0.417 0.676733
## smokeHxUnknown     -0.0534333  0.9479691  0.0705105  0.0842372  -0.634 0.525872
## smokeHxEver        -0.0001398  0.9998603  0.0498816  0.0146130  -0.010 0.992369
## disadv              0.0194098  1.0195994  0.0456339  0.0706709   0.275 0.783585
## site02R             0.1394609  1.1496539  0.2585339  0.1014674   1.374 0.169305
## site03R            -0.4106006  0.6632518  0.2676881  0.1151425  -3.566 0.000362
## site04R            -1.0460758  0.3513137  0.2762447  0.1211479  -8.635  < 2e-16
## site05R            -1.4455201  0.2356235  0.2840259  0.0568837 -25.412  < 2e-16
## site06R            -0.9516495  0.3861036  0.2742355  0.0962182  -9.891  < 2e-16
## site07R            -1.3756514  0.2526750  0.2514679  0.1862686  -7.385 1.52e-13
## site09R            -0.7416434  0.4763305  0.2761663  0.0673538 -11.011  < 2e-16
## site1              -1.5544149  0.2113130  0.2274027  0.0719897 -21.592  < 2e-16
## site101            -1.1335593  0.3218855  0.2282521  0.2010436  -5.638 1.72e-08
## site102            -0.6376679  0.5285236  0.2272866  0.1546209  -4.124 3.72e-05
## site103            -0.3352224  0.7151790  0.2227826  0.1082563  -3.097 0.001958
## site104            -0.6595980  0.5170592  0.2289029  0.1281763  -5.146 2.66e-07
## site105            -1.0161273  0.3619941  0.2257368  0.1881986  -5.399 6.69e-08
## site106            -1.3566176  0.2575304  0.2275374  0.1704373  -7.960 1.73e-15
## site107            -0.7176227  0.4879108  0.2481484  0.2087767  -3.437 0.000588
## site108            -0.6447068  0.5248164  0.2468598  0.2847586  -2.264 0.023571
## site10R            -0.8852539  0.4126094  0.2970140  0.1527895  -5.794 6.88e-09
## site11R            -1.1588617  0.3138432  0.2466684  0.1069997 -10.831  < 2e-16
## site12R            -0.6236159  0.5360028  0.2572520  0.0365245 -17.074  < 2e-16
## site13R            -1.4270846  0.2400076  0.2468433  0.1774411  -8.043 8.80e-16
## site14R            -0.7228767  0.4853540  0.3994901  0.1160683  -6.228 4.72e-10
## site15R            -0.8985221  0.4071710  0.2764243  0.0811261 -11.076  < 2e-16
## site16R            -0.5418222  0.5816873  0.2681892  0.0413544 -13.102  < 2e-16
## site17R            -1.0283076  0.3576117  0.2811913  0.0646268 -15.911  < 2e-16
## site18R            -1.4565403  0.2330411  0.2657102  0.2465883  -5.907 3.49e-09
## site19R            -0.2920568  0.7467261  0.2870797  0.0632030  -4.621 3.82e-06
## site20R            -0.9920734  0.3708070  0.2922973  0.0635554 -15.610  < 2e-16
## site21R            -1.3475211  0.2598837  0.2586169  0.0881674 -15.284  < 2e-16
## site22R            -0.9842263  0.3737283  0.2532427  0.0724626 -13.583  < 2e-16
## site23R            -1.0841697  0.3381825  0.2564349  0.0770784 -14.066  < 2e-16
## site24R            -0.4111608  0.6628803  0.2539202  0.0319734 -12.859  < 2e-16
## site25R            -0.4514899  0.6366789  0.2568174  0.1107135  -4.078 4.54e-05
## site26R            -1.4838953  0.2267527  0.2720764  0.1871254  -7.930 2.19e-15
## site27R            -1.1160514  0.3275707  0.3734863  0.1067333 -10.456  < 2e-16
## site28R            -1.2792304  0.2782514  0.2729314  0.0779668 -16.407  < 2e-16
## site29R            -1.1244751  0.3248229  0.3397708  0.1540943  -7.297 2.94e-13
## site30R            -1.2954233  0.2737820  0.2618878  0.1858099  -6.972 3.13e-12
## site31R            -1.0238891  0.3591953  0.2758866  0.0619911 -16.517  < 2e-16
## site32R            -0.7423082  0.4760139  0.2781651  0.1166662  -6.363 1.98e-10
## site33R            -1.3058681  0.2709372  0.2646573  0.1301099 -10.037  < 2e-16
## site34R            -1.0127968  0.3632018  0.2538071  0.1276968  -7.931 2.17e-15
## site35R            -0.5995747  0.5490451  0.2704420  0.0450046 -13.323  < 2e-16
## site36R            -1.3918254  0.2486211  0.2612833  0.0821215 -16.948  < 2e-16
## site37R            -1.3688891  0.2543894  0.2631513  0.1521625  -8.996  < 2e-16
## site38R            -1.1675748  0.3111206  0.2678772  0.0735074 -15.884  < 2e-16
## site39R            -0.4805553  0.6184399  0.2939729  0.0747571  -6.428 1.29e-10
## site40R            -0.7233672  0.4851160  0.2719657  0.1854047  -3.902 9.56e-05
## site41R            -1.0144742  0.3625930  0.2569589  0.0935923 -10.839  < 2e-16
## site42R            -0.9088722  0.4029784  0.2810688  0.0448728 -20.254  < 2e-16
##                       
## SO4                ***
## dx_yr              ** 
## age_dx             ** 
## sexF               *  
## dich_RaceNon-White    
## smokeHxFormer      ***
## smokeHxAlways         
## smokeHxUnknown        
## smokeHxEver           
## disadv                
## site02R               
## site03R            ***
## site04R            ***
## site05R            ***
## site06R            ***
## site07R            ***
## site09R            ***
## site1              ***
## site101            ***
## site102            ***
## site103            ** 
## site104            ***
## site105            ***
## site106            ***
## site107            ***
## site108            *  
## site10R            ***
## site11R            ***
## site12R            ***
## site13R            ***
## site14R            ***
## site15R            ***
## site16R            ***
## site17R            ***
## site18R            ***
## site19R            ***
## site20R            ***
## site21R            ***
## site22R            ***
## site23R            ***
## site24R            ***
## site25R            ***
## site26R            ***
## site27R            ***
## site28R            ***
## site29R            ***
## site30R            ***
## site31R            ***
## site32R            ***
## site33R            ***
## site34R            ***
## site35R            ***
## site36R            ***
## site37R            ***
## site38R            ***
## site39R            ***
## site40R            ***
## site41R            ***
## site42R            ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SO4                   2.1141     0.4730    1.8066    2.4739
## dx_yr                 1.4101     0.7092    1.1031    1.8025
## age_dx                1.0092     0.9909    1.0024    1.0159
## sexF                  0.8730     1.1455    0.7717    0.9876
## dich_RaceNon-White    1.0175     0.9828    0.9488    1.0911
## smokeHxFormer         1.1291     0.8856    1.0788    1.1818
## smokeHxAlways         1.0254     0.9752    0.9114    1.1536
## smokeHxUnknown        0.9480     1.0549    0.8037    1.1181
## smokeHxEver           0.9999     1.0001    0.9716    1.0289
## disadv                1.0196     0.9808    0.8877    1.1711
## site02R               1.1497     0.8698    0.9423    1.4026
## site03R               0.6633     1.5077    0.5293    0.8312
## site04R               0.3513     2.8465    0.2771    0.4455
## site05R               0.2356     4.2441    0.2108    0.2634
## site06R               0.3861     2.5900    0.3197    0.4662
## site07R               0.2527     3.9577    0.1754    0.3640
## site09R               0.4763     2.0994    0.4174    0.5436
## site1                 0.2113     4.7323    0.1835    0.2433
## site101               0.3219     3.1067    0.2171    0.4773
## site102               0.5285     1.8921    0.3903    0.7156
## site103               0.7152     1.3983    0.5785    0.8842
## site104               0.5171     1.9340    0.4022    0.6647
## site105               0.3620     2.7625    0.2503    0.5235
## site106               0.2575     3.8830    0.1844    0.3597
## site107               0.4879     2.0496    0.3241    0.7346
## site108               0.5248     1.9054    0.3003    0.9171
## site10R               0.4126     2.4236    0.3058    0.5567
## site11R               0.3138     3.1863    0.2545    0.3871
## site12R               0.5360     1.8657    0.4990    0.5758
## site13R               0.2400     4.1665    0.1695    0.3398
## site14R               0.4854     2.0604    0.3866    0.6093
## site15R               0.4072     2.4560    0.3473    0.4773
## site16R               0.5817     1.7191    0.5364    0.6308
## site17R               0.3576     2.7963    0.3151    0.4059
## site18R               0.2330     4.2911    0.1437    0.3779
## site19R               0.7467     1.3392    0.6597    0.8452
## site20R               0.3708     2.6968    0.3274    0.4200
## site21R               0.2599     3.8479    0.2186    0.3089
## site22R               0.3737     2.6757    0.3242    0.4308
## site23R               0.3382     2.9570    0.2908    0.3933
## site24R               0.6629     1.5086    0.6226    0.7058
## site25R               0.6367     1.5707    0.5125    0.7910
## site26R               0.2268     4.4101    0.1571    0.3272
## site27R               0.3276     3.0528    0.2657    0.4038
## site28R               0.2783     3.5939    0.2388    0.3242
## site29R               0.3248     3.0786    0.2401    0.4394
## site30R               0.2738     3.6525    0.1902    0.3941
## site31R               0.3592     2.7840    0.3181    0.4056
## site32R               0.4760     2.1008    0.3787    0.5983
## site33R               0.2709     3.6909    0.2100    0.3496
## site34R               0.3632     2.7533    0.2828    0.4665
## site35R               0.5490     1.8213    0.5027    0.5997
## site36R               0.2486     4.0222    0.2117    0.2920
## site37R               0.2544     3.9310    0.1888    0.3428
## site38R               0.3111     3.2142    0.2694    0.3593
## site39R               0.6184     1.6170    0.5342    0.7160
## site40R               0.4851     2.0614    0.3373    0.6977
## site41R               0.3626     2.7579    0.3018    0.4356
## site42R               0.4030     2.4815    0.3691    0.4400
## 
## Concordance= 0.686  (se = 0.064 )
## Likelihood ratio test= 3187  on 59 df,   p=<2e-16
## Wald test            = 258.2  on 59 df,   p=<2e-16
## Score (logrank) test = 2590  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

88.2.2 SO4 Per IQR

summary(All$SO4)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##  0.1354  0.6529  1.4104  1.6701  2.1343  6.6991     688

IQR(All$SO4, na.rm=T)

## [1] 1.481359

# Will use the 5yr pre-censoring IQR (0.9653843), not this one
All <- All %>% mutate(SO4_IQR = SO4/0.9653843)

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SO4_IQR + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4_IQR + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)  robust se       z Pr(>|z|)
## SO4_IQR             0.7227055  2.0599990  0.0221052  0.0774219   9.335  < 2e-16
## dx_yr               0.3436630  1.4101033  0.0070602  0.1252699   2.743 0.006081
## age_dx              0.0091231  1.0091649  0.0011770  0.0034138   2.672 0.007531
## sexF               -0.1358048  0.8730131  0.0267003  0.0629412  -2.158 0.030955
## dich_RaceNon-White  0.0173323  1.0174833  0.0368078  0.0356521   0.486 0.626861
## smokeHxFormer       0.1214572  1.1291410  0.0328996  0.0232463   5.225 1.74e-07
## smokeHxAlways       0.0250676  1.0253845  0.0748671  0.0601250   0.417 0.676733
## smokeHxUnknown     -0.0534333  0.9479691  0.0705105  0.0842372  -0.634 0.525872
## smokeHxEver        -0.0001398  0.9998603  0.0498816  0.0146130  -0.010 0.992369
## disadv              0.0194098  1.0195994  0.0456339  0.0706709   0.275 0.783585
## site02R             0.1394609  1.1496539  0.2585339  0.1014674   1.374 0.169305
## site03R            -0.4106006  0.6632518  0.2676881  0.1151425  -3.566 0.000362
## site04R            -1.0460758  0.3513137  0.2762447  0.1211479  -8.635  < 2e-16
## site05R            -1.4455201  0.2356235  0.2840259  0.0568837 -25.412  < 2e-16
## site06R            -0.9516495  0.3861036  0.2742355  0.0962182  -9.891  < 2e-16
## site07R            -1.3756514  0.2526750  0.2514679  0.1862686  -7.385 1.52e-13
## site09R            -0.7416434  0.4763305  0.2761663  0.0673538 -11.011  < 2e-16
## site1              -1.5544149  0.2113130  0.2274027  0.0719897 -21.592  < 2e-16
## site101            -1.1335593  0.3218855  0.2282521  0.2010436  -5.638 1.72e-08
## site102            -0.6376679  0.5285236  0.2272866  0.1546209  -4.124 3.72e-05
## site103            -0.3352224  0.7151790  0.2227826  0.1082563  -3.097 0.001958
## site104            -0.6595980  0.5170592  0.2289029  0.1281763  -5.146 2.66e-07
## site105            -1.0161273  0.3619941  0.2257368  0.1881986  -5.399 6.69e-08
## site106            -1.3566176  0.2575304  0.2275374  0.1704373  -7.960 1.73e-15
## site107            -0.7176227  0.4879108  0.2481484  0.2087767  -3.437 0.000588
## site108            -0.6447068  0.5248164  0.2468598  0.2847586  -2.264 0.023571
## site10R            -0.8852539  0.4126094  0.2970140  0.1527895  -5.794 6.88e-09
## site11R            -1.1588617  0.3138432  0.2466684  0.1069997 -10.831  < 2e-16
## site12R            -0.6236159  0.5360028  0.2572520  0.0365245 -17.074  < 2e-16
## site13R            -1.4270846  0.2400076  0.2468433  0.1774411  -8.043 8.80e-16
## site14R            -0.7228767  0.4853540  0.3994901  0.1160683  -6.228 4.72e-10
## site15R            -0.8985221  0.4071710  0.2764243  0.0811261 -11.076  < 2e-16
## site16R            -0.5418222  0.5816873  0.2681892  0.0413544 -13.102  < 2e-16
## site17R            -1.0283076  0.3576117  0.2811913  0.0646268 -15.911  < 2e-16
## site18R            -1.4565403  0.2330411  0.2657102  0.2465883  -5.907 3.49e-09
## site19R            -0.2920568  0.7467261  0.2870797  0.0632030  -4.621 3.82e-06
## site20R            -0.9920734  0.3708070  0.2922973  0.0635554 -15.610  < 2e-16
## site21R            -1.3475211  0.2598837  0.2586169  0.0881674 -15.284  < 2e-16
## site22R            -0.9842263  0.3737283  0.2532427  0.0724626 -13.583  < 2e-16
## site23R            -1.0841697  0.3381825  0.2564349  0.0770784 -14.066  < 2e-16
## site24R            -0.4111608  0.6628803  0.2539202  0.0319734 -12.859  < 2e-16
## site25R            -0.4514899  0.6366789  0.2568174  0.1107135  -4.078 4.54e-05
## site26R            -1.4838953  0.2267527  0.2720764  0.1871254  -7.930 2.19e-15
## site27R            -1.1160514  0.3275707  0.3734863  0.1067333 -10.456  < 2e-16
## site28R            -1.2792304  0.2782514  0.2729314  0.0779668 -16.407  < 2e-16
## site29R            -1.1244751  0.3248229  0.3397708  0.1540943  -7.297 2.94e-13
## site30R            -1.2954233  0.2737820  0.2618878  0.1858099  -6.972 3.13e-12
## site31R            -1.0238891  0.3591953  0.2758866  0.0619911 -16.517  < 2e-16
## site32R            -0.7423082  0.4760139  0.2781651  0.1166662  -6.363 1.98e-10
## site33R            -1.3058681  0.2709372  0.2646573  0.1301099 -10.037  < 2e-16
## site34R            -1.0127968  0.3632018  0.2538071  0.1276968  -7.931 2.17e-15
## site35R            -0.5995747  0.5490451  0.2704420  0.0450046 -13.323  < 2e-16
## site36R            -1.3918254  0.2486211  0.2612833  0.0821215 -16.948  < 2e-16
## site37R            -1.3688891  0.2543894  0.2631513  0.1521625  -8.996  < 2e-16
## site38R            -1.1675748  0.3111206  0.2678772  0.0735074 -15.884  < 2e-16
## site39R            -0.4805553  0.6184399  0.2939729  0.0747571  -6.428 1.29e-10
## site40R            -0.7233672  0.4851160  0.2719657  0.1854047  -3.902 9.56e-05
## site41R            -1.0144742  0.3625930  0.2569589  0.0935923 -10.839  < 2e-16
## site42R            -0.9088722  0.4029784  0.2810688  0.0448728 -20.254  < 2e-16
##                       
## SO4_IQR            ***
## dx_yr              ** 
## age_dx             ** 
## sexF               *  
## dich_RaceNon-White    
## smokeHxFormer      ***
## smokeHxAlways         
## smokeHxUnknown        
## smokeHxEver           
## disadv                
## site02R               
## site03R            ***
## site04R            ***
## site05R            ***
## site06R            ***
## site07R            ***
## site09R            ***
## site1              ***
## site101            ***
## site102            ***
## site103            ** 
## site104            ***
## site105            ***
## site106            ***
## site107            ***
## site108            *  
## site10R            ***
## site11R            ***
## site12R            ***
## site13R            ***
## site14R            ***
## site15R            ***
## site16R            ***
## site17R            ***
## site18R            ***
## site19R            ***
## site20R            ***
## site21R            ***
## site22R            ***
## site23R            ***
## site24R            ***
## site25R            ***
## site26R            ***
## site27R            ***
## site28R            ***
## site29R            ***
## site30R            ***
## site31R            ***
## site32R            ***
## site33R            ***
## site34R            ***
## site35R            ***
## site36R            ***
## site37R            ***
## site38R            ***
## site39R            ***
## site40R            ***
## site41R            ***
## site42R            ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SO4_IQR               2.0600     0.4854    1.7700    2.3976
## dx_yr                 1.4101     0.7092    1.1031    1.8025
## age_dx                1.0092     0.9909    1.0024    1.0159
## sexF                  0.8730     1.1455    0.7717    0.9876
## dich_RaceNon-White    1.0175     0.9828    0.9488    1.0911
## smokeHxFormer         1.1291     0.8856    1.0788    1.1818
## smokeHxAlways         1.0254     0.9752    0.9114    1.1536
## smokeHxUnknown        0.9480     1.0549    0.8037    1.1181
## smokeHxEver           0.9999     1.0001    0.9716    1.0289
## disadv                1.0196     0.9808    0.8877    1.1711
## site02R               1.1497     0.8698    0.9423    1.4026
## site03R               0.6633     1.5077    0.5293    0.8312
## site04R               0.3513     2.8465    0.2771    0.4455
## site05R               0.2356     4.2441    0.2108    0.2634
## site06R               0.3861     2.5900    0.3197    0.4662
## site07R               0.2527     3.9577    0.1754    0.3640
## site09R               0.4763     2.0994    0.4174    0.5436
## site1                 0.2113     4.7323    0.1835    0.2433
## site101               0.3219     3.1067    0.2171    0.4773
## site102               0.5285     1.8921    0.3903    0.7156
## site103               0.7152     1.3983    0.5785    0.8842
## site104               0.5171     1.9340    0.4022    0.6647
## site105               0.3620     2.7625    0.2503    0.5235
## site106               0.2575     3.8830    0.1844    0.3597
## site107               0.4879     2.0496    0.3241    0.7346
## site108               0.5248     1.9054    0.3003    0.9171
## site10R               0.4126     2.4236    0.3058    0.5567
## site11R               0.3138     3.1863    0.2545    0.3871
## site12R               0.5360     1.8657    0.4990    0.5758
## site13R               0.2400     4.1665    0.1695    0.3398
## site14R               0.4854     2.0604    0.3866    0.6093
## site15R               0.4072     2.4560    0.3473    0.4773
## site16R               0.5817     1.7191    0.5364    0.6308
## site17R               0.3576     2.7963    0.3151    0.4059
## site18R               0.2330     4.2911    0.1437    0.3779
## site19R               0.7467     1.3392    0.6597    0.8452
## site20R               0.3708     2.6968    0.3274    0.4200
## site21R               0.2599     3.8479    0.2186    0.3089
## site22R               0.3737     2.6757    0.3242    0.4308
## site23R               0.3382     2.9570    0.2908    0.3933
## site24R               0.6629     1.5086    0.6226    0.7058
## site25R               0.6367     1.5707    0.5125    0.7910
## site26R               0.2268     4.4101    0.1571    0.3272
## site27R               0.3276     3.0528    0.2657    0.4038
## site28R               0.2783     3.5939    0.2388    0.3242
## site29R               0.3248     3.0786    0.2401    0.4394
## site30R               0.2738     3.6525    0.1902    0.3941
## site31R               0.3592     2.7840    0.3181    0.4056
## site32R               0.4760     2.1008    0.3787    0.5983
## site33R               0.2709     3.6909    0.2100    0.3496
## site34R               0.3632     2.7533    0.2828    0.4665
## site35R               0.5490     1.8213    0.5027    0.5997
## site36R               0.2486     4.0222    0.2117    0.2920
## site37R               0.2544     3.9310    0.1888    0.3428
## site38R               0.3111     3.2142    0.2694    0.3593
## site39R               0.6184     1.6170    0.5342    0.7160
## site40R               0.4851     2.0614    0.3373    0.6977
## site41R               0.3626     2.7579    0.3018    0.4356
## site42R               0.4030     2.4815    0.3691    0.4400
## 
## Concordance= 0.686  (se = 0.064 )
## Likelihood ratio test= 3187  on 59 df,   p=<2e-16
## Wald test            = 258.2  on 59 df,   p=<2e-16
## Score (logrank) test = 2590  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

So this indicates that there is a HR of 2.05 per IQR increase in SO4 as compared with a HR of 2.11 per 1ug/m3 increase in SO4.

88.2.3 SO4 Variable HR Spline Models

Base model

#First need to make dataframe that only includes patients with a value for event
Allx <- All %>% filter(!is.na(SO4) & !is.na(deadORtx) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(cohort) & !is.na(site) & SO4<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$SO4, df=3) + Allx$dx_yr + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$SO4, exp(predicted$fit), type="n")
lines(sm.spline(Allx$SO4, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$SO4, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$SO4, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

Complete model

#First need to make dataframe that only includes patients with time_DeathTxCensor
Allx <- All %>% filter(!is.na(SO4) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(deadORtx) & !is.na(age_dx) & !is.na(sex) & !is.na(smokeHx) & !is.na(dich_Race) & !is.na(disadv) & !is.na(site) & SO4<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$SO4, df=3) + Allx$dx_yr + Allx$age_dx + Allx$sex + Allx$smokeHx + Allx$dich_Race + Allx$disadv + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$SO4, exp(predicted$fit), type="n")
lines(sm.spline(Allx$SO4, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$SO4, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$SO4, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

88.3 NO3

88.3.1 NO3 Continuous Models

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ NO3 + dx_yr + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3 + dx_yr + 
##     site, data = All, id = ID, cluster = cohort)
## 
##   n= 335367, number of events= 6459 
##    (1981 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se      z Pr(>|z|)    
## NO3     -0.021509  0.978721  0.052338  0.106715 -0.202  0.84027    
## dx_yr    0.218105  1.243717  0.005658  0.155595  1.402  0.16099    
## site02R  0.492259  1.636008  0.257029  0.226539  2.173  0.02978 *  
## site03R -0.211901  0.809044  0.263642  0.120585 -1.757  0.07887 .  
## site04R -0.161510  0.850858  0.278215  0.208475 -0.775  0.43851    
## site05R -0.169848  0.843793  0.278414  0.070846 -2.397  0.01651 *  
## site06R  0.101265  1.106570  0.267093  0.073260  1.382  0.16689    
## site07R -0.397046  0.672303  0.251428  0.295040 -1.346  0.17839    
## site09R -0.062169  0.939724  0.270074  0.123793 -0.502  0.61553    
## site1   -0.104759  0.900542  0.217478  0.158766 -0.660  0.50936    
## site101 -0.359880  0.697760  0.222071  0.303268 -1.187  0.23536    
## site102 -0.445886  0.640257  0.220307  0.229350 -1.944  0.05188 .  
## site103 -0.164343  0.848451  0.215497  0.153011 -1.074  0.28280    
## site104 -0.340593  0.711349  0.221274  0.170711 -1.995  0.04603 *  
## site105 -0.313069  0.731200  0.218308  0.289843 -1.080  0.28008    
## site106 -0.353308  0.702361  0.222203  0.276973 -1.276  0.20210    
## site107  0.011366  1.011431  0.241472  0.327069  0.035  0.97228    
## site108 -0.123338  0.883965  0.240910  0.463727 -0.266  0.79026    
## site10R -0.250208  0.778639  0.289740  0.154203 -1.623  0.10468    
## site11R -0.076466  0.926384  0.238429  0.100103 -0.764  0.44494    
## site12R -0.097380  0.907211  0.255903  0.094582 -1.030  0.30320    
## site13R -0.370677  0.690267  0.241987  0.235064 -1.577  0.11481    
## site14R  0.135180  1.144743  0.398037  0.097536  1.386  0.16576    
## site15R  0.024692  1.024999  0.270647  0.058370  0.423  0.67227    
## site16R  0.157296  1.170342  0.265234  0.023727  6.629 3.37e-11 ***
## site17R -0.049507  0.951699  0.278571  0.106378 -0.465  0.64166    
## site18R -0.436036  0.646595  0.259487  0.294133 -1.482  0.13822    
## site19R -0.090277  0.913678  0.282857  0.051374 -1.757  0.07888 .  
## site20R -0.207490  0.812621  0.289565  0.087297 -2.377  0.01746 *  
## site21R -0.281511  0.754642  0.252426  0.053142 -5.297 1.17e-07 ***
## site22R -0.020363  0.979842  0.251067  0.138222 -0.147  0.88288    
## site23R -0.144319  0.865612  0.248754  0.073816 -1.955  0.05057 .  
## site24R  0.040953  1.041803  0.254214  0.044151  0.928  0.35364    
## site25R -0.093084  0.911117  0.256622  0.220754 -0.422  0.67327    
## site26R -0.353691  0.702092  0.265547  0.146109 -2.421  0.01549 *  
## site27R -0.147198  0.863123  0.370411  0.124580 -1.182  0.23738    
## site28R -0.031776  0.968724  0.269727  0.143530 -0.221  0.82479    
## site29R -0.324466  0.722913  0.335573  0.154830 -2.096  0.03611 *  
## site30R -0.250547  0.778375  0.265243  0.320008 -0.783  0.43366    
## site31R -0.191453  0.825758  0.270714  0.071143 -2.691  0.00712 ** 
## site32R -0.418893  0.657775  0.268177  0.158640 -2.641  0.00828 ** 
## site33R -0.287528  0.750115  0.257624  0.102005 -2.819  0.00482 ** 
## site34R -0.207735  0.812422  0.248795  0.153014 -1.358  0.17458    
## site35R -0.071443  0.931050  0.264357  0.067425 -1.060  0.28933    
## site36R -0.198459  0.819993  0.259579  0.138344 -1.435  0.15142    
## site37R -0.223170  0.799979  0.261880  0.247253 -0.903  0.36674    
## site38R -0.278244  0.757112  0.261180  0.070099 -3.969 7.21e-05 ***
## site39R -0.248342  0.780093  0.291634  0.157763 -1.574  0.11545    
## site40R  0.082887  1.086419  0.285975  0.433906  0.191  0.84851    
## site41R -0.248450  0.780009  0.254794  0.175858 -1.413  0.15772    
## site42R  0.150365  1.162259  0.277853  0.063031  2.386  0.01705 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## NO3        0.9787     1.0217    0.7940    1.2064
## dx_yr      1.2437     0.8040    0.9168    1.6872
## site02R    1.6360     0.6112    1.0494    2.5504
## site03R    0.8090     1.2360    0.6387    1.0247
## site04R    0.8509     1.1753    0.5655    1.2803
## site05R    0.8438     1.1851    0.7344    0.9695
## site06R    1.1066     0.9037    0.9586    1.2774
## site07R    0.6723     1.4874    0.3771    1.1987
## site09R    0.9397     1.0641    0.7373    1.1978
## site1      0.9005     1.1104    0.6597    1.2293
## site101    0.6978     1.4332    0.3851    1.2643
## site102    0.6403     1.5619    0.4084    1.0036
## site103    0.8485     1.1786    0.6286    1.1452
## site104    0.7113     1.4058    0.5091    0.9940
## site105    0.7312     1.3676    0.4143    1.2905
## site106    0.7024     1.4238    0.4081    1.2087
## site107    1.0114     0.9887    0.5328    1.9202
## site108    0.8840     1.1313    0.3562    2.1936
## site10R    0.7786     1.2843    0.5755    1.0534
## site11R    0.9264     1.0795    0.7613    1.1272
## site12R    0.9072     1.1023    0.7537    1.0920
## site13R    0.6903     1.4487    0.4354    1.0942
## site14R    1.1447     0.8736    0.9456    1.3859
## site15R    1.0250     0.9756    0.9142    1.1492
## site16R    1.1703     0.8545    1.1172    1.2261
## site17R    0.9517     1.0508    0.7726    1.1723
## site18R    0.6466     1.5466    0.3633    1.1508
## site19R    0.9137     1.0945    0.8262    1.0105
## site20R    0.8126     1.2306    0.6848    0.9643
## site21R    0.7546     1.3251    0.6800    0.8375
## site22R    0.9798     1.0206    0.7473    1.2847
## site23R    0.8656     1.1553    0.7490    1.0004
## site24R    1.0418     0.9599    0.9554    1.1360
## site25R    0.9111     1.0976    0.5911    1.4044
## site26R    0.7021     1.4243    0.5273    0.9349
## site27R    0.8631     1.1586    0.6761    1.1018
## site28R    0.9687     1.0323    0.7312    1.2834
## site29R    0.7229     1.3833    0.5337    0.9792
## site30R    0.7784     1.2847    0.4157    1.4574
## site31R    0.8258     1.2110    0.7183    0.9493
## site32R    0.6578     1.5203    0.4820    0.8977
## site33R    0.7501     1.3331    0.6142    0.9161
## site34R    0.8124     1.2309    0.6019    1.0965
## site35R    0.9310     1.0741    0.8158    1.0626
## site36R    0.8200     1.2195    0.6252    1.0754
## site37R    0.8000     1.2500    0.4927    1.2988
## site38R    0.7571     1.3208    0.6599    0.8686
## site39R    0.7801     1.2819    0.5726    1.0628
## site40R    1.0864     0.9205    0.4641    2.5430
## site41R    0.7800     1.2820    0.5526    1.1010
## site42R    1.1623     0.8604    1.0272    1.3151
## 
## Concordance= 0.673  (se = 0.067 )
## Likelihood ratio test= 2368  on 51 df,   p=<2e-16
## Wald test            = 13.79  on 51 df,   p=1
## Score (logrank) test = 1909  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NO3 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se      z Pr(>|z|)    
## NO3                -0.008507  0.991529  0.053139  0.114125 -0.075 0.940582    
## dx_yr               0.215262  1.240186  0.005823  0.159008  1.354 0.175808    
## age_dx              0.009958  1.010007  0.001177  0.003977  2.504 0.012295 *  
## sexF               -0.141777  0.867815  0.026687  0.061665 -2.299 0.021497 *  
## dich_RaceNon-White  0.005301  1.005315  0.036768  0.036088  0.147 0.883226    
## smokeHxFormer       0.105323  1.111070  0.032876  0.054535  1.931 0.053446 .  
## smokeHxAlways       0.047988  1.049158  0.074817  0.023847  2.012 0.044187 *  
## smokeHxUnknown      0.061439  1.063365  0.070542  0.038548  1.594 0.110973    
## smokeHxEver        -0.006549  0.993472  0.049914  0.012427 -0.527 0.598170    
## disadv              0.005159  1.005172  0.045473  0.047703  0.108 0.913885    
## site02R             0.465406  1.592661  0.261527  0.228983  2.032 0.042104 *  
## site03R            -0.201467  0.817531  0.267979  0.122596 -1.643 0.100312    
## site04R            -0.129377  0.878643  0.282780  0.233514 -0.554 0.579550    
## site05R            -0.149882  0.860809  0.284467  0.060909 -2.461 0.013865 *  
## site06R             0.066749  1.069028  0.273000  0.102670  0.650 0.515605    
## site07R            -0.448004  0.638902  0.256082  0.291040 -1.539 0.123727    
## site09R            -0.113719  0.892509  0.276333  0.108040 -1.053 0.292540    
## site1              -0.169597  0.844005  0.225101  0.142391 -1.191 0.233628    
## site101            -0.380467  0.683542  0.229063  0.286450 -1.328 0.184106    
## site102            -0.486239  0.614935  0.227491  0.209996 -2.315 0.020588 *  
## site103            -0.196553  0.821558  0.222663  0.119326 -1.647 0.099519 .  
## site104            -0.395691  0.673215  0.228217  0.132492 -2.987 0.002822 ** 
## site105            -0.365534  0.693826  0.225846  0.266183 -1.373 0.169677    
## site106            -0.460711  0.630835  0.229561  0.242425 -1.900 0.057377 .  
## site107            -0.051238  0.950053  0.248162  0.299594 -0.171 0.864205    
## site108            -0.191710  0.825546  0.247357  0.433422 -0.442 0.658259    
## site10R            -0.297053  0.743005  0.296541  0.143386 -2.072 0.038293 *  
## site11R            -0.137943  0.871148  0.244817  0.088189 -1.564 0.117776    
## site12R            -0.143489  0.866330  0.260613  0.092045 -1.559 0.119018    
## site13R            -0.452695  0.635912  0.246686  0.226880 -1.995 0.046010 *  
## site14R             0.137957  1.147926  0.400917  0.103911  1.328 0.184295    
## site15R            -0.038092  0.962625  0.275257  0.049682 -0.767 0.443251    
## site16R             0.173553  1.189524  0.269432  0.039229  4.424 9.69e-06 ***
## site17R            -0.014330  0.985773  0.282993  0.122206 -0.117 0.906655    
## site18R            -0.474896  0.621950  0.264106  0.276282 -1.719 0.085636 .  
## site19R            -0.128958  0.879011  0.287121  0.036472 -3.536 0.000406 ***
## site20R            -0.231198  0.793582  0.293837  0.092520 -2.499 0.012458 *  
## site21R            -0.321606  0.724984  0.256910  0.042597 -7.550 4.35e-14 ***
## site22R            -0.080358  0.922786  0.255870  0.135310 -0.594 0.552593    
## site23R            -0.172176  0.841831  0.255531  0.054615 -3.153 0.001619 ** 
## site24R             0.057019  1.058675  0.259085  0.067855  0.840 0.400738    
## site25R            -0.130761  0.877428  0.263101  0.227869 -0.574 0.566074    
## site26R            -0.407310  0.665438  0.270116  0.135366 -3.009 0.002622 ** 
## site27R            -0.136854  0.872098  0.373494  0.130238 -1.051 0.293350    
## site28R             0.030488  1.030957  0.274244  0.156947  0.194 0.845978    
## site29R            -0.361106  0.696905  0.339663  0.161800 -2.232 0.025628 *  
## site30R            -0.316432  0.728744  0.270691  0.320053 -0.989 0.322816    
## site31R            -0.188192  0.828455  0.274849  0.071589 -2.629 0.008569 ** 
## site32R            -0.415412  0.660069  0.278171  0.146648 -2.833 0.004615 ** 
## site33R            -0.327836  0.720481  0.263074  0.079675 -4.115 3.88e-05 ***
## site34R            -0.252667  0.776726  0.253962  0.148488 -1.702 0.088832 .  
## site35R            -0.124583  0.882865  0.270182  0.050038 -2.490 0.012782 *  
## site36R            -0.208284  0.811976  0.264184  0.139149 -1.497 0.134433    
## site37R            -0.275480  0.759208  0.266388  0.247827 -1.112 0.266319    
## site38R            -0.300539  0.740419  0.266683  0.066902 -4.492 7.05e-06 ***
## site39R            -0.305335  0.736876  0.295957  0.144781 -2.109 0.034949 *  
## site40R             0.009934  1.009983  0.292066  0.429334  0.023 0.981541    
## site41R            -0.264493  0.767595  0.259202  0.177673 -1.489 0.136579    
## site42R             0.143293  1.154068  0.281946  0.070257  2.040 0.041396 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NO3                   0.9915     1.0085    0.7928    1.2401
## dx_yr                 1.2402     0.8063    0.9081    1.6937
## age_dx                1.0100     0.9901    1.0022    1.0179
## sexF                  0.8678     1.1523    0.7690    0.9793
## dich_RaceNon-White    1.0053     0.9947    0.9367    1.0790
## smokeHxFormer         1.1111     0.9000    0.9984    1.2364
## smokeHxAlways         1.0492     0.9531    1.0012    1.0994
## smokeHxUnknown        1.0634     0.9404    0.9860    1.1468
## smokeHxEver           0.9935     1.0066    0.9696    1.0180
## disadv                1.0052     0.9949    0.9155    1.1037
## site02R               1.5927     0.6279    1.0167    2.4948
## site03R               0.8175     1.2232    0.6429    1.0396
## site04R               0.8786     1.1381    0.5560    1.3886
## site05R               0.8608     1.1617    0.7639    0.9700
## site06R               1.0690     0.9354    0.8742    1.3073
## site07R               0.6389     1.5652    0.3612    1.1302
## site09R               0.8925     1.1204    0.7222    1.1030
## site1                 0.8440     1.1848    0.6385    1.1157
## site101               0.6835     1.4630    0.3899    1.1984
## site102               0.6149     1.6262    0.4075    0.9281
## site103               0.8216     1.2172    0.6502    1.0380
## site104               0.6732     1.4854    0.5192    0.8728
## site105               0.6938     1.4413    0.4118    1.1690
## site106               0.6308     1.5852    0.3922    1.0145
## site107               0.9501     1.0526    0.5281    1.7091
## site108               0.8255     1.2113    0.3530    1.9305
## site10R               0.7430     1.3459    0.5610    0.9841
## site11R               0.8711     1.1479    0.7329    1.0355
## site12R               0.8663     1.1543    0.7233    1.0376
## site13R               0.6359     1.5725    0.4076    0.9920
## site14R               1.1479     0.8711    0.9364    1.4072
## site15R               0.9626     1.0388    0.8733    1.0611
## site16R               1.1895     0.8407    1.1015    1.2846
## site17R               0.9858     1.0144    0.7758    1.2526
## site18R               0.6219     1.6078    0.3619    1.0689
## site19R               0.8790     1.1376    0.8184    0.9441
## site20R               0.7936     1.2601    0.6620    0.9514
## site21R               0.7250     1.3793    0.6669    0.7881
## site22R               0.9228     1.0837    0.7078    1.2030
## site23R               0.8418     1.1879    0.7564    0.9369
## site24R               1.0587     0.9446    0.9268    1.2093
## site25R               0.8774     1.1397    0.5614    1.3714
## site26R               0.6654     1.5028    0.5104    0.8676
## site27R               0.8721     1.1467    0.6756    1.1257
## site28R               1.0310     0.9700    0.7580    1.4023
## site29R               0.6969     1.4349    0.5075    0.9570
## site30R               0.7287     1.3722    0.3892    1.3646
## site31R               0.8285     1.2071    0.7200    0.9532
## site32R               0.6601     1.5150    0.4952    0.8799
## site33R               0.7205     1.3880    0.6163    0.8423
## site34R               0.7767     1.2875    0.5806    1.0391
## site35R               0.8829     1.1327    0.8004    0.9738
## site36R               0.8120     1.2316    0.6182    1.0666
## site37R               0.7592     1.3172    0.4671    1.2340
## site38R               0.7404     1.3506    0.6494    0.8442
## site39R               0.7369     1.3571    0.5548    0.9787
## site40R               1.0100     0.9901    0.4354    2.3430
## site41R               0.7676     1.3028    0.5419    1.0873
## site42R               1.1541     0.8665    1.0056    1.3244
## 
## Concordance= 0.676  (se = 0.064 )
## Likelihood ratio test= 2535  on 59 df,   p=<2e-16
## Wald test            = 12.32  on 59 df,   p=1
## Score (logrank) test = 2067  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

88.3.2 NO3 Per IQR

summary(All$NO3)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##  0.0134  0.4777  0.8362  0.8433  1.1149  4.7595     688

IQR(All$NO3, na.rm=T)

## [1] 0.6372749

# Will use the 5yr pre-censoring IQR (0.5319728), not this one
All <- All %>% mutate(NO3_IQR = NO3/0.5319728)

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NO3_IQR + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3_IQR + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se      z Pr(>|z|)    
## NO3_IQR            -0.004525  0.995485  0.028268  0.060712 -0.075 0.940582    
## dx_yr               0.215262  1.240186  0.005823  0.159008  1.354 0.175808    
## age_dx              0.009958  1.010007  0.001177  0.003977  2.504 0.012295 *  
## sexF               -0.141777  0.867815  0.026687  0.061665 -2.299 0.021497 *  
## dich_RaceNon-White  0.005301  1.005315  0.036768  0.036088  0.147 0.883226    
## smokeHxFormer       0.105323  1.111070  0.032876  0.054535  1.931 0.053446 .  
## smokeHxAlways       0.047988  1.049158  0.074817  0.023847  2.012 0.044187 *  
## smokeHxUnknown      0.061439  1.063365  0.070542  0.038548  1.594 0.110973    
## smokeHxEver        -0.006549  0.993472  0.049914  0.012427 -0.527 0.598170    
## disadv              0.005159  1.005172  0.045473  0.047703  0.108 0.913885    
## site02R             0.465406  1.592661  0.261527  0.228983  2.032 0.042104 *  
## site03R            -0.201467  0.817531  0.267979  0.122596 -1.643 0.100312    
## site04R            -0.129377  0.878643  0.282780  0.233514 -0.554 0.579550    
## site05R            -0.149882  0.860809  0.284467  0.060909 -2.461 0.013865 *  
## site06R             0.066749  1.069028  0.273000  0.102670  0.650 0.515605    
## site07R            -0.448004  0.638902  0.256082  0.291040 -1.539 0.123727    
## site09R            -0.113719  0.892509  0.276333  0.108040 -1.053 0.292540    
## site1              -0.169597  0.844005  0.225101  0.142391 -1.191 0.233628    
## site101            -0.380467  0.683542  0.229063  0.286450 -1.328 0.184106    
## site102            -0.486239  0.614935  0.227491  0.209996 -2.315 0.020588 *  
## site103            -0.196553  0.821558  0.222663  0.119326 -1.647 0.099519 .  
## site104            -0.395691  0.673215  0.228217  0.132492 -2.987 0.002822 ** 
## site105            -0.365534  0.693826  0.225846  0.266183 -1.373 0.169677    
## site106            -0.460711  0.630835  0.229561  0.242425 -1.900 0.057377 .  
## site107            -0.051238  0.950053  0.248162  0.299594 -0.171 0.864205    
## site108            -0.191710  0.825546  0.247357  0.433422 -0.442 0.658259    
## site10R            -0.297053  0.743005  0.296541  0.143386 -2.072 0.038293 *  
## site11R            -0.137943  0.871148  0.244817  0.088189 -1.564 0.117776    
## site12R            -0.143489  0.866330  0.260613  0.092045 -1.559 0.119018    
## site13R            -0.452695  0.635912  0.246686  0.226880 -1.995 0.046010 *  
## site14R             0.137957  1.147926  0.400917  0.103911  1.328 0.184295    
## site15R            -0.038092  0.962625  0.275257  0.049682 -0.767 0.443251    
## site16R             0.173553  1.189524  0.269432  0.039229  4.424 9.69e-06 ***
## site17R            -0.014330  0.985773  0.282993  0.122206 -0.117 0.906655    
## site18R            -0.474896  0.621950  0.264106  0.276282 -1.719 0.085636 .  
## site19R            -0.128958  0.879011  0.287121  0.036472 -3.536 0.000406 ***
## site20R            -0.231198  0.793582  0.293837  0.092520 -2.499 0.012458 *  
## site21R            -0.321606  0.724984  0.256910  0.042597 -7.550 4.35e-14 ***
## site22R            -0.080358  0.922786  0.255870  0.135310 -0.594 0.552593    
## site23R            -0.172176  0.841831  0.255531  0.054615 -3.153 0.001619 ** 
## site24R             0.057019  1.058675  0.259085  0.067855  0.840 0.400738    
## site25R            -0.130761  0.877428  0.263101  0.227869 -0.574 0.566074    
## site26R            -0.407310  0.665438  0.270116  0.135366 -3.009 0.002622 ** 
## site27R            -0.136854  0.872098  0.373494  0.130238 -1.051 0.293350    
## site28R             0.030488  1.030957  0.274244  0.156947  0.194 0.845978    
## site29R            -0.361106  0.696905  0.339663  0.161800 -2.232 0.025628 *  
## site30R            -0.316432  0.728744  0.270691  0.320053 -0.989 0.322816    
## site31R            -0.188192  0.828455  0.274849  0.071589 -2.629 0.008569 ** 
## site32R            -0.415412  0.660069  0.278171  0.146648 -2.833 0.004615 ** 
## site33R            -0.327836  0.720481  0.263074  0.079675 -4.115 3.88e-05 ***
## site34R            -0.252667  0.776726  0.253962  0.148488 -1.702 0.088832 .  
## site35R            -0.124583  0.882865  0.270182  0.050038 -2.490 0.012782 *  
## site36R            -0.208284  0.811976  0.264184  0.139149 -1.497 0.134433    
## site37R            -0.275480  0.759208  0.266388  0.247827 -1.112 0.266319    
## site38R            -0.300539  0.740419  0.266683  0.066902 -4.492 7.05e-06 ***
## site39R            -0.305335  0.736876  0.295957  0.144781 -2.109 0.034949 *  
## site40R             0.009934  1.009983  0.292066  0.429334  0.023 0.981541    
## site41R            -0.264493  0.767595  0.259202  0.177673 -1.489 0.136579    
## site42R             0.143293  1.154068  0.281946  0.070257  2.040 0.041396 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NO3_IQR               0.9955     1.0045    0.8838    1.1213
## dx_yr                 1.2402     0.8063    0.9081    1.6937
## age_dx                1.0100     0.9901    1.0022    1.0179
## sexF                  0.8678     1.1523    0.7690    0.9793
## dich_RaceNon-White    1.0053     0.9947    0.9367    1.0790
## smokeHxFormer         1.1111     0.9000    0.9984    1.2364
## smokeHxAlways         1.0492     0.9531    1.0012    1.0994
## smokeHxUnknown        1.0634     0.9404    0.9860    1.1468
## smokeHxEver           0.9935     1.0066    0.9696    1.0180
## disadv                1.0052     0.9949    0.9155    1.1037
## site02R               1.5927     0.6279    1.0167    2.4948
## site03R               0.8175     1.2232    0.6429    1.0396
## site04R               0.8786     1.1381    0.5560    1.3886
## site05R               0.8608     1.1617    0.7639    0.9700
## site06R               1.0690     0.9354    0.8742    1.3073
## site07R               0.6389     1.5652    0.3612    1.1302
## site09R               0.8925     1.1204    0.7222    1.1030
## site1                 0.8440     1.1848    0.6385    1.1157
## site101               0.6835     1.4630    0.3899    1.1984
## site102               0.6149     1.6262    0.4075    0.9281
## site103               0.8216     1.2172    0.6502    1.0380
## site104               0.6732     1.4854    0.5192    0.8728
## site105               0.6938     1.4413    0.4118    1.1690
## site106               0.6308     1.5852    0.3922    1.0145
## site107               0.9501     1.0526    0.5281    1.7091
## site108               0.8255     1.2113    0.3530    1.9305
## site10R               0.7430     1.3459    0.5610    0.9841
## site11R               0.8711     1.1479    0.7329    1.0355
## site12R               0.8663     1.1543    0.7233    1.0376
## site13R               0.6359     1.5725    0.4076    0.9920
## site14R               1.1479     0.8711    0.9364    1.4072
## site15R               0.9626     1.0388    0.8733    1.0611
## site16R               1.1895     0.8407    1.1015    1.2846
## site17R               0.9858     1.0144    0.7758    1.2526
## site18R               0.6219     1.6078    0.3619    1.0689
## site19R               0.8790     1.1376    0.8184    0.9441
## site20R               0.7936     1.2601    0.6620    0.9514
## site21R               0.7250     1.3793    0.6669    0.7881
## site22R               0.9228     1.0837    0.7078    1.2030
## site23R               0.8418     1.1879    0.7564    0.9369
## site24R               1.0587     0.9446    0.9268    1.2093
## site25R               0.8774     1.1397    0.5614    1.3714
## site26R               0.6654     1.5028    0.5104    0.8676
## site27R               0.8721     1.1467    0.6756    1.1257
## site28R               1.0310     0.9700    0.7580    1.4023
## site29R               0.6969     1.4349    0.5075    0.9570
## site30R               0.7287     1.3722    0.3892    1.3646
## site31R               0.8285     1.2071    0.7200    0.9532
## site32R               0.6601     1.5150    0.4952    0.8799
## site33R               0.7205     1.3880    0.6163    0.8423
## site34R               0.7767     1.2875    0.5806    1.0391
## site35R               0.8829     1.1327    0.8004    0.9738
## site36R               0.8120     1.2316    0.6182    1.0666
## site37R               0.7592     1.3172    0.4671    1.2340
## site38R               0.7404     1.3506    0.6494    0.8442
## site39R               0.7369     1.3571    0.5548    0.9787
## site40R               1.0100     0.9901    0.4354    2.3430
## site41R               0.7676     1.3028    0.5419    1.0873
## site42R               1.1541     0.8665    1.0056    1.3244
## 
## Concordance= 0.676  (se = 0.064 )
## Likelihood ratio test= 2535  on 59 df,   p=<2e-16
## Wald test            = 12.32  on 59 df,   p=1
## Score (logrank) test = 2067  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

So this indicates that there is a HR of 0.99 per IQR increase in NO3 as compared with a HR of 0.99 per 1ug/m3 increase in NO3.

88.3.3 NO3 Variable HR Spline Models

Base model

#First need to make dataframe that only includes patients with a value for event
Allx <- All %>% filter(!is.na(NO3) & !is.na(deadORtx) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(cohort) & !is.na(site) & NO3<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$NO3, df=3) + Allx$dx_yr + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$NO3, exp(predicted$fit), type="n")
lines(sm.spline(Allx$NO3, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$NO3, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$NO3, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

Complete model

#First need to make dataframe that only includes patients with time_DeathTxCensor
Allx <- All %>% filter(!is.na(NO3) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(deadORtx) & !is.na(age_dx) & !is.na(sex) & !is.na(smokeHx) & !is.na(dich_Race) & !is.na(disadv) & !is.na(site) & NO3<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$NO3, df=3) + Allx$dx_yr + Allx$age_dx + Allx$sex + Allx$smokeHx + Allx$dich_Race + Allx$disadv + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$NO3, exp(predicted$fit), type="n")
lines(sm.spline(Allx$NO3, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$NO3, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$NO3, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

88.4 NH4

88.4.1 NH4 Continuous Models

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ NH4 + dx_yr + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4 + dx_yr + 
##     site, data = All, id = ID, cluster = cohort)
## 
##   n= 335367, number of events= 6459 
##    (1981 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se      z Pr(>|z|)    
## NH4      1.671858  5.322045  0.078078  0.789905  2.117 0.034300 *  
## dx_yr    0.349796  1.418779  0.008574  0.182352  1.918 0.055080 .  
## site02R  0.015702  1.015826  0.254970  0.318107  0.049 0.960631    
## site03R -0.407128  0.665559  0.263394  0.177984 -2.287 0.022170 *  
## site04R -1.149122  0.316915  0.274386  0.426498 -2.694 0.007053 ** 
## site05R -1.256124  0.284756  0.279734  0.407766 -3.081 0.002067 ** 
## site06R -0.514479  0.597812  0.268206  0.152451 -3.375 0.000739 ***
## site07R -1.364070  0.255618  0.249266  0.541877 -2.517 0.011826 *  
## site09R -0.582663  0.558409  0.270301  0.233206 -2.498 0.012472 *  
## site1   -1.270590  0.280666  0.222627  0.516141 -2.462 0.013828 *  
## site101 -1.053338  0.348771  0.222486  0.478951 -2.199 0.027859 *  
## site102 -0.635929  0.529443  0.220224  0.229006 -2.777 0.005488 ** 
## site103 -0.207757  0.812405  0.215538  0.159307 -1.304 0.192190    
## site104 -0.412959  0.661690  0.221325  0.173762 -2.377 0.017474 *  
## site105 -0.891426  0.410071  0.218940  0.412888 -2.159 0.030850 *  
## site106 -1.230615  0.292113  0.222255  0.510115 -2.412 0.015847 *  
## site107 -0.583263  0.558075  0.242190  0.442516 -1.318 0.187484    
## site108 -0.553043  0.575197  0.240919  0.476934 -1.160 0.246220    
## site10R -0.316924  0.728386  0.289772  0.163726 -1.936 0.052905 .  
## site11R -0.450807  0.637114  0.238993  0.200705 -2.246 0.024697 *  
## site12R -0.694032  0.499558  0.253733  0.225550 -3.077 0.002090 ** 
## site13R -1.094469  0.334717  0.242708  0.418865 -2.613 0.008977 ** 
## site14R -0.781498  0.457720  0.398287  0.220497 -3.544 0.000394 ***
## site15R -0.206152  0.813710  0.270828  0.115281 -1.788 0.073736 .  
## site16R -0.475886  0.621335  0.264706  0.177679 -2.678 0.007399 ** 
## site17R -0.897244  0.407692  0.277932  0.352543 -2.545 0.010926 *  
## site18R -0.852169  0.426489  0.260123  0.382293 -2.229 0.025807 *  
## site19R -0.072818  0.929770  0.282761  0.061768 -1.179 0.238437    
## site20R -0.835086  0.433837  0.288569  0.242828 -3.439 0.000584 ***
## site21R -0.702296  0.495447  0.252989  0.171608 -4.092 4.27e-05 ***
## site22R -0.918628  0.399066  0.250665  0.363929 -2.524 0.011596 *  
## site23R -0.671166  0.511112  0.249262  0.226980 -2.957 0.003107 ** 
## site24R -0.627308  0.534027  0.250465  0.179728 -3.490 0.000482 ***
## site25R -0.792201  0.452847  0.253472  0.464712 -1.705 0.088247 .  
## site26R -0.914951  0.400536  0.266795  0.366534 -2.496 0.012552 *  
## site27R -0.919708  0.398635  0.370966  0.370671 -2.481 0.013094 *  
## site28R -1.165767  0.311684  0.270463  0.469547 -2.483 0.013037 *  
## site29R -0.873819  0.417355  0.335746  0.289868 -3.015 0.002574 ** 
## site30R -1.327709  0.265084  0.259744  0.540400 -2.457 0.014014 *  
## site31R -0.658991  0.517373  0.271421  0.206701 -3.188 0.001432 ** 
## site32R -0.650542  0.521763  0.268253  0.228661 -2.845 0.004441 ** 
## site33R -0.825359  0.438078  0.258706  0.309252 -2.669 0.007610 ** 
## site34R -0.861920  0.422350  0.249408  0.344729 -2.500 0.012409 *  
## site35R -0.367377  0.692549  0.264593  0.128078 -2.868 0.004126 ** 
## site36R -1.373731  0.253161  0.259917  0.480864 -2.857 0.004279 ** 
## site37R -1.341163  0.261541  0.261838  0.534750 -2.508 0.012141 *  
## site38R -0.672289  0.510539  0.261710  0.177859 -3.780 0.000157 ***
## site39R -0.616723  0.539710  0.290216  0.233117 -2.646 0.008156 ** 
## site40R -1.302602  0.271824  0.273611  0.788581 -1.652 0.098569 .  
## site41R -0.923654  0.397066  0.253462  0.308677 -2.992 0.002769 ** 
## site42R -0.742172  0.476079  0.278205  0.318720 -2.329 0.019880 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## NH4        5.3220     0.1879   1.13165   25.0292
## dx_yr      1.4188     0.7048   0.99242    2.0283
## site02R    1.0158     0.9844   0.54456    1.8949
## site03R    0.6656     1.5025   0.46956    0.9434
## site04R    0.3169     3.1554   0.13737    0.7311
## site05R    0.2848     3.5118   0.12805    0.6332
## site06R    0.5978     1.6728   0.44340    0.8060
## site07R    0.2556     3.9121   0.08838    0.7393
## site09R    0.5584     1.7908   0.35355    0.8820
## site1      0.2807     3.5630   0.10206    0.7718
## site101    0.3488     2.8672   0.13641    0.8917
## site102    0.5294     1.8888   0.33798    0.8294
## site103    0.8124     1.2309   0.59452    1.1101
## site104    0.6617     1.5113   0.47070    0.9302
## site105    0.4101     2.4386   0.18256    0.9211
## site106    0.2921     3.4233   0.10748    0.7939
## site107    0.5581     1.7919   0.23443    1.3285
## site108    0.5752     1.7385   0.22587    1.4648
## site10R    0.7284     1.3729   0.52844    1.0040
## site11R    0.6371     1.5696   0.42991    0.9442
## site12R    0.4996     2.0018   0.32107    0.7773
## site13R    0.3347     2.9876   0.14728    0.7607
## site14R    0.4577     2.1847   0.29711    0.7052
## site15R    0.8137     1.2289   0.64914    1.0200
## site16R    0.6213     1.6094   0.43862    0.8802
## site17R    0.4077     2.4528   0.20429    0.8136
## site18R    0.4265     2.3447   0.20160    0.9022
## site19R    0.9298     1.0755   0.82376    1.0494
## site20R    0.4338     2.3050   0.26955    0.6983
## site21R    0.4954     2.0184   0.35394    0.6935
## site22R    0.3991     2.5058   0.19555    0.8144
## site23R    0.5111     1.9565   0.32757    0.7975
## site24R    0.5340     1.8726   0.37547    0.7595
## site25R    0.4528     2.2083   0.18213    1.1259
## site26R    0.4005     2.4967   0.19528    0.8216
## site27R    0.3986     2.5086   0.19278    0.8243
## site28R    0.3117     3.2084   0.12418    0.7823
## site29R    0.4174     2.3960   0.23647    0.7366
## site30R    0.2651     3.7724   0.09192    0.7645
## site31R    0.5174     1.9328   0.34503    0.7758
## site32R    0.5218     1.9166   0.33330    0.8168
## site33R    0.4381     2.2827   0.23895    0.8031
## site34R    0.4224     2.3677   0.21490    0.8301
## site35R    0.6925     1.4439   0.53880    0.8902
## site36R    0.2532     3.9501   0.09865    0.6497
## site37R    0.2615     3.8235   0.09170    0.7460
## site38R    0.5105     1.9587   0.36028    0.7235
## site39R    0.5397     1.8528   0.34177    0.8523
## site40R    0.2718     3.6789   0.05795    1.2751
## site41R    0.3971     2.5185   0.21683    0.7271
## site42R    0.4761     2.1005   0.25491    0.8891
## 
## Concordance= 0.674  (se = 0.065 )
## Likelihood ratio test= 2797  on 51 df,   p=<2e-16
## Wald test            = 6.45  on 51 df,   p=1
## Score (logrank) test = 2202  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NH4 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)  robust se      z Pr(>|z|)
## NH4                 1.7096909  5.5272525  0.0794455  0.7768404  2.201 0.027748
## dx_yr               0.3493734  1.4181786  0.0087555  0.1836184  1.903 0.057078
## age_dx              0.0097076  1.0097549  0.0011748  0.0036607  2.652 0.008006
## sexF               -0.1376702  0.8713860  0.0266937  0.0592741 -2.323 0.020201
## dich_RaceNon-White  0.0106919  1.0107492  0.0367194  0.0274967  0.389 0.697393
## smokeHxFormer       0.0914987  1.0958154  0.0328697  0.0395747  2.312 0.020775
## smokeHxAlways      -0.0001908  0.9998093  0.0748577  0.0395371 -0.005 0.996150
## smokeHxUnknown     -0.0520493  0.9492820  0.0706821  0.0257670 -2.020 0.043383
## smokeHxEver        -0.0095266  0.9905187  0.0499679  0.0146788 -0.649 0.516336
## disadv              0.0151891  1.0153050  0.0453374  0.0571111  0.266 0.790273
## site02R             0.0061263  1.0061451  0.2593874  0.3018780  0.020 0.983809
## site03R            -0.3749202  0.6873441  0.2677471  0.1632482 -2.297 0.021640
## site04R            -1.1024364  0.3320611  0.2785575  0.4081305 -2.701 0.006909
## site05R            -1.2228944  0.2943769  0.2856961  0.3595665 -3.401 0.000671
## site06R            -0.5365883  0.5847398  0.2740228  0.1053477 -5.094 3.52e-07
## site07R            -1.3979246  0.2471093  0.2537142  0.5124788 -2.728 0.006376
## site09R            -0.6220183  0.5368598  0.2765829  0.2086381 -2.981 0.002870
## site1              -1.3106019  0.2696577  0.2295537  0.4672318 -2.805 0.005031
## site101            -1.0604423  0.3463026  0.2291769  0.4387317 -2.417 0.015646
## site102            -0.6465331  0.5238588  0.2273540  0.2015462 -3.208 0.001337
## site103            -0.2181307  0.8040203  0.2227175  0.1227607 -1.777 0.075588
## site104            -0.4467383  0.6397113  0.2282395  0.1333780 -3.349 0.000810
## site105            -0.9235732  0.3970976  0.2262424  0.3682674 -2.508 0.012146
## site106            -1.3192127  0.2673457  0.2291752  0.4574226 -2.884 0.003926
## site107            -0.6166431  0.5397533  0.2486007  0.4006009 -1.539 0.123732
## site108            -0.6006326  0.5484645  0.2471343  0.4386309 -1.369 0.170895
## site10R            -0.3528529  0.7026806  0.2965086  0.1504501 -2.345 0.019011
## site11R            -0.5063852  0.6026701  0.2453713  0.1757924 -2.881 0.003969
## site12R            -0.7235460  0.4850293  0.2582603  0.2019557 -3.583 0.000340
## site13R            -1.1645825  0.3120529  0.2472749  0.3981161 -2.925 0.003442
## site14R            -0.7547489  0.4701287  0.4011653  0.1838000 -4.106 4.02e-05
## site15R            -0.2595795  0.7713759  0.2754597  0.0950880 -2.730 0.006336
## site16R            -0.4464280  0.6399098  0.2688001  0.1568797 -2.846 0.004432
## site17R            -0.8541173  0.4256588  0.2821947  0.3256450 -2.623 0.008720
## site18R            -0.8809265  0.4143988  0.2646959  0.3512191 -2.508 0.012135
## site19R            -0.0944777  0.9098480  0.2870442  0.0402124 -2.349 0.018800
## site20R            -0.8446219  0.4297198  0.2927504  0.2217039 -3.810 0.000139
## site21R            -0.7299419  0.4819370  0.2574301  0.1489144 -4.902 9.50e-07
## site22R            -0.9623246  0.3820038  0.2552191  0.3391971 -2.837 0.004553
## site23R            -0.6833593  0.5049180  0.2559961  0.1952208 -3.500 0.000464
## site24R            -0.5870177  0.5559829  0.2552275  0.1526914 -3.844 0.000121
## site25R            -0.8361038  0.4333959  0.2597864  0.4574819 -1.828 0.067606
## site26R            -0.9617908  0.3822078  0.2712678  0.3435822 -2.799 0.005121
## site27R            -0.8947363  0.4087154  0.3739394  0.3489923 -2.564 0.010354
## site28R            -1.0948183  0.3346004  0.2749568  0.4349577 -2.517 0.011834
## site29R            -0.8943912  0.4088564  0.3397935  0.2852147 -3.136 0.001714
## site30R            -1.3838622  0.2506088  0.2647684  0.5178907 -2.672 0.007538
## site31R            -0.6411507  0.5266860  0.2755237  0.1871218 -3.426 0.000612
## site32R            -0.6397453  0.5274267  0.2782498  0.2005939 -3.189 0.001426
## site33R            -0.8386735  0.4322836  0.2640502  0.2631108 -3.188 0.001435
## site34R            -0.8935119  0.4092161  0.2544624  0.3242382 -2.756 0.005856
## site35R            -0.4066764  0.6658596  0.2703584  0.1046750 -3.885 0.000102
## site36R            -1.3746689  0.2529233  0.2643669  0.4480875 -3.068 0.002156
## site37R            -1.3791152  0.2518013  0.2661370  0.5086673 -2.711 0.006703
## site38R            -0.6795467  0.5068467  0.2671766  0.1557021 -4.364 1.27e-05
## site39R            -0.6566515  0.5185849  0.2944146  0.2047666 -3.207 0.001342
## site40R            -1.3711046  0.2538264  0.2790531  0.7610127 -1.802 0.071595
## site41R            -0.9240745  0.3968986  0.2577492  0.2854500 -3.237 0.001207
## site42R            -0.7375335  0.4782922  0.2821641  0.2947501 -2.502 0.012341
##                       
## NH4                *  
## dx_yr              .  
## age_dx             ** 
## sexF               *  
## dich_RaceNon-White    
## smokeHxFormer      *  
## smokeHxAlways         
## smokeHxUnknown     *  
## smokeHxEver           
## disadv                
## site02R               
## site03R            *  
## site04R            ** 
## site05R            ***
## site06R            ***
## site07R            ** 
## site09R            ** 
## site1              ** 
## site101            *  
## site102            ** 
## site103            .  
## site104            ***
## site105            *  
## site106            ** 
## site107               
## site108               
## site10R            *  
## site11R            ** 
## site12R            ***
## site13R            ** 
## site14R            ***
## site15R            ** 
## site16R            ** 
## site17R            ** 
## site18R            *  
## site19R            *  
## site20R            ***
## site21R            ***
## site22R            ** 
## site23R            ***
## site24R            ***
## site25R            .  
## site26R            ** 
## site27R            *  
## site28R            *  
## site29R            ** 
## site30R            ** 
## site31R            ***
## site32R            ** 
## site33R            ** 
## site34R            ** 
## site35R            ***
## site36R            ** 
## site37R            ** 
## site38R            ***
## site39R            ** 
## site40R            .  
## site41R            ** 
## site42R            *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NH4                   5.5273     0.1809   1.20576   25.3371
## dx_yr                 1.4182     0.7051   0.98954    2.0325
## age_dx                1.0098     0.9903   1.00254    1.0170
## sexF                  0.8714     1.1476   0.77581    0.9787
## dich_RaceNon-White    1.0107     0.9894   0.95772    1.0667
## smokeHxFormer         1.0958     0.9126   1.01403    1.1842
## smokeHxAlways         0.9998     1.0002   0.92526    1.0804
## smokeHxUnknown        0.9493     1.0534   0.90253    0.9985
## smokeHxEver           0.9905     1.0096   0.96243    1.0194
## disadv                1.0153     0.9849   0.90779    1.1356
## site02R               1.0061     0.9939   0.55680    1.8181
## site03R               0.6873     1.4549   0.49913    0.9465
## site04R               0.3321     3.0115   0.14922    0.7390
## site05R               0.2944     3.3970   0.14549    0.5956
## site06R               0.5847     1.7102   0.47565    0.7188
## site07R               0.2471     4.0468   0.09050    0.6747
## site09R               0.5369     1.8627   0.35667    0.8081
## site1                 0.2697     3.7084   0.10792    0.6738
## site101               0.3463     2.8876   0.14656    0.8183
## site102               0.5239     1.9089   0.35290    0.7776
## site103               0.8040     1.2437   0.63208    1.0227
## site104               0.6397     1.5632   0.49255    0.8308
## site105               0.3971     2.5183   0.19294    0.8173
## site106               0.2673     3.7405   0.10907    0.6553
## site107               0.5398     1.8527   0.24615    1.1836
## site108               0.5485     1.8233   0.23216    1.2957
## site10R               0.7027     1.4231   0.52323    0.9437
## site11R               0.6027     1.6593   0.42702    0.8506
## site12R               0.4850     2.0617   0.32648    0.7206
## site13R               0.3121     3.2046   0.14301    0.6809
## site14R               0.4701     2.1271   0.32792    0.6740
## site15R               0.7714     1.2964   0.64022    0.9294
## site16R               0.6399     1.5627   0.47052    0.8703
## site17R               0.4257     2.3493   0.22484    0.8058
## site18R               0.4144     2.4131   0.20819    0.8249
## site19R               0.9098     1.0991   0.84089    0.9845
## site20R               0.4297     2.3271   0.27827    0.6636
## site21R               0.4819     2.0750   0.35994    0.6453
## site22R               0.3820     2.6178   0.19649    0.7427
## site23R               0.5049     1.9805   0.34439    0.7403
## site24R               0.5560     1.7986   0.41218    0.7500
## site25R               0.4334     2.3074   0.17680    1.0624
## site26R               0.3822     2.6164   0.19491    0.7495
## site27R               0.4087     2.4467   0.20623    0.8100
## site28R               0.3346     2.9886   0.14266    0.7848
## site29R               0.4089     2.4458   0.23377    0.7151
## site30R               0.2506     3.9903   0.09082    0.6916
## site31R               0.5267     1.8987   0.36498    0.7600
## site32R               0.5274     1.8960   0.35597    0.7815
## site33R               0.4323     2.3133   0.25811    0.7240
## site34R               0.4092     2.4437   0.21675    0.7726
## site35R               0.6659     1.5018   0.54235    0.8175
## site36R               0.2529     3.9538   0.10509    0.6087
## site37R               0.2518     3.9714   0.09291    0.6824
## site38R               0.5068     1.9730   0.37354    0.6877
## site39R               0.5186     1.9283   0.34715    0.7747
## site40R               0.2538     3.9397   0.05712    1.1280
## site41R               0.3969     2.5195   0.22683    0.6945
## site42R               0.4783     2.0908   0.26841    0.8523
## 
## Concordance= 0.68  (se = 0.064 )
## Likelihood ratio test= 2967  on 59 df,   p=<2e-16
## Wald test            = 8.84  on 59 df,   p=1
## Score (logrank) test = 2343  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

88.4.2 NH4 Per IQR

summary(All$NH4)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##  0.0000  0.2151  0.5016  0.6227  0.8336  2.6082     688

IQR(All$NH4, na.rm=T)

## [1] 0.6185003

# Will use the 5yr pre-censoring IQR (0.3400484), not this one
All <- All %>% mutate(NH4_IQR = NH4/0.3400484)

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NH4_IQR + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4_IQR + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)  robust se      z Pr(>|z|)
## NH4_IQR             0.5813776  1.7885006  0.0270153  0.2641634  2.201 0.027748
## dx_yr               0.3493734  1.4181786  0.0087555  0.1836184  1.903 0.057078
## age_dx              0.0097076  1.0097549  0.0011748  0.0036607  2.652 0.008006
## sexF               -0.1376702  0.8713860  0.0266937  0.0592741 -2.323 0.020201
## dich_RaceNon-White  0.0106919  1.0107492  0.0367194  0.0274967  0.389 0.697393
## smokeHxFormer       0.0914987  1.0958154  0.0328697  0.0395747  2.312 0.020775
## smokeHxAlways      -0.0001908  0.9998093  0.0748577  0.0395371 -0.005 0.996150
## smokeHxUnknown     -0.0520493  0.9492820  0.0706821  0.0257670 -2.020 0.043383
## smokeHxEver        -0.0095266  0.9905187  0.0499679  0.0146788 -0.649 0.516336
## disadv              0.0151891  1.0153050  0.0453374  0.0571111  0.266 0.790273
## site02R             0.0061263  1.0061451  0.2593874  0.3018780  0.020 0.983809
## site03R            -0.3749202  0.6873441  0.2677471  0.1632482 -2.297 0.021640
## site04R            -1.1024364  0.3320611  0.2785575  0.4081305 -2.701 0.006909
## site05R            -1.2228944  0.2943769  0.2856961  0.3595665 -3.401 0.000671
## site06R            -0.5365883  0.5847398  0.2740228  0.1053477 -5.094 3.52e-07
## site07R            -1.3979246  0.2471093  0.2537142  0.5124788 -2.728 0.006376
## site09R            -0.6220183  0.5368598  0.2765829  0.2086381 -2.981 0.002870
## site1              -1.3106019  0.2696577  0.2295537  0.4672318 -2.805 0.005031
## site101            -1.0604423  0.3463026  0.2291769  0.4387317 -2.417 0.015646
## site102            -0.6465331  0.5238588  0.2273540  0.2015462 -3.208 0.001337
## site103            -0.2181307  0.8040203  0.2227175  0.1227607 -1.777 0.075588
## site104            -0.4467383  0.6397113  0.2282395  0.1333780 -3.349 0.000810
## site105            -0.9235732  0.3970976  0.2262424  0.3682674 -2.508 0.012146
## site106            -1.3192127  0.2673457  0.2291752  0.4574226 -2.884 0.003926
## site107            -0.6166431  0.5397533  0.2486007  0.4006009 -1.539 0.123732
## site108            -0.6006326  0.5484645  0.2471343  0.4386309 -1.369 0.170895
## site10R            -0.3528529  0.7026806  0.2965086  0.1504501 -2.345 0.019011
## site11R            -0.5063852  0.6026701  0.2453713  0.1757924 -2.881 0.003969
## site12R            -0.7235460  0.4850293  0.2582603  0.2019557 -3.583 0.000340
## site13R            -1.1645825  0.3120529  0.2472749  0.3981161 -2.925 0.003442
## site14R            -0.7547489  0.4701287  0.4011653  0.1838000 -4.106 4.02e-05
## site15R            -0.2595795  0.7713759  0.2754597  0.0950880 -2.730 0.006336
## site16R            -0.4464280  0.6399098  0.2688001  0.1568797 -2.846 0.004432
## site17R            -0.8541173  0.4256588  0.2821947  0.3256450 -2.623 0.008720
## site18R            -0.8809265  0.4143988  0.2646959  0.3512191 -2.508 0.012135
## site19R            -0.0944777  0.9098480  0.2870442  0.0402124 -2.349 0.018800
## site20R            -0.8446219  0.4297198  0.2927504  0.2217039 -3.810 0.000139
## site21R            -0.7299419  0.4819370  0.2574301  0.1489144 -4.902 9.50e-07
## site22R            -0.9623246  0.3820038  0.2552191  0.3391971 -2.837 0.004553
## site23R            -0.6833593  0.5049180  0.2559961  0.1952208 -3.500 0.000464
## site24R            -0.5870177  0.5559829  0.2552275  0.1526914 -3.844 0.000121
## site25R            -0.8361038  0.4333959  0.2597864  0.4574819 -1.828 0.067606
## site26R            -0.9617908  0.3822078  0.2712678  0.3435822 -2.799 0.005121
## site27R            -0.8947363  0.4087154  0.3739394  0.3489923 -2.564 0.010354
## site28R            -1.0948183  0.3346004  0.2749568  0.4349577 -2.517 0.011834
## site29R            -0.8943912  0.4088564  0.3397935  0.2852147 -3.136 0.001714
## site30R            -1.3838622  0.2506088  0.2647684  0.5178907 -2.672 0.007538
## site31R            -0.6411507  0.5266860  0.2755237  0.1871218 -3.426 0.000612
## site32R            -0.6397453  0.5274267  0.2782498  0.2005939 -3.189 0.001426
## site33R            -0.8386735  0.4322836  0.2640502  0.2631108 -3.188 0.001435
## site34R            -0.8935119  0.4092161  0.2544624  0.3242382 -2.756 0.005856
## site35R            -0.4066764  0.6658596  0.2703584  0.1046750 -3.885 0.000102
## site36R            -1.3746689  0.2529233  0.2643669  0.4480875 -3.068 0.002156
## site37R            -1.3791152  0.2518013  0.2661370  0.5086673 -2.711 0.006703
## site38R            -0.6795467  0.5068467  0.2671766  0.1557021 -4.364 1.27e-05
## site39R            -0.6566515  0.5185849  0.2944146  0.2047666 -3.207 0.001342
## site40R            -1.3711046  0.2538264  0.2790531  0.7610127 -1.802 0.071595
## site41R            -0.9240745  0.3968986  0.2577492  0.2854500 -3.237 0.001207
## site42R            -0.7375335  0.4782922  0.2821641  0.2947501 -2.502 0.012341
##                       
## NH4_IQR            *  
## dx_yr              .  
## age_dx             ** 
## sexF               *  
## dich_RaceNon-White    
## smokeHxFormer      *  
## smokeHxAlways         
## smokeHxUnknown     *  
## smokeHxEver           
## disadv                
## site02R               
## site03R            *  
## site04R            ** 
## site05R            ***
## site06R            ***
## site07R            ** 
## site09R            ** 
## site1              ** 
## site101            *  
## site102            ** 
## site103            .  
## site104            ***
## site105            *  
## site106            ** 
## site107               
## site108               
## site10R            *  
## site11R            ** 
## site12R            ***
## site13R            ** 
## site14R            ***
## site15R            ** 
## site16R            ** 
## site17R            ** 
## site18R            *  
## site19R            *  
## site20R            ***
## site21R            ***
## site22R            ** 
## site23R            ***
## site24R            ***
## site25R            .  
## site26R            ** 
## site27R            *  
## site28R            *  
## site29R            ** 
## site30R            ** 
## site31R            ***
## site32R            ** 
## site33R            ** 
## site34R            ** 
## site35R            ***
## site36R            ** 
## site37R            ** 
## site38R            ***
## site39R            ** 
## site40R            .  
## site41R            ** 
## site42R            *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NH4_IQR               1.7885     0.5591   1.06569    3.0015
## dx_yr                 1.4182     0.7051   0.98954    2.0325
## age_dx                1.0098     0.9903   1.00254    1.0170
## sexF                  0.8714     1.1476   0.77581    0.9787
## dich_RaceNon-White    1.0107     0.9894   0.95772    1.0667
## smokeHxFormer         1.0958     0.9126   1.01403    1.1842
## smokeHxAlways         0.9998     1.0002   0.92526    1.0804
## smokeHxUnknown        0.9493     1.0534   0.90253    0.9985
## smokeHxEver           0.9905     1.0096   0.96243    1.0194
## disadv                1.0153     0.9849   0.90779    1.1356
## site02R               1.0061     0.9939   0.55680    1.8181
## site03R               0.6873     1.4549   0.49913    0.9465
## site04R               0.3321     3.0115   0.14922    0.7390
## site05R               0.2944     3.3970   0.14549    0.5956
## site06R               0.5847     1.7102   0.47565    0.7188
## site07R               0.2471     4.0468   0.09050    0.6747
## site09R               0.5369     1.8627   0.35667    0.8081
## site1                 0.2697     3.7084   0.10792    0.6738
## site101               0.3463     2.8876   0.14656    0.8183
## site102               0.5239     1.9089   0.35290    0.7776
## site103               0.8040     1.2437   0.63208    1.0227
## site104               0.6397     1.5632   0.49255    0.8308
## site105               0.3971     2.5183   0.19294    0.8173
## site106               0.2673     3.7405   0.10907    0.6553
## site107               0.5398     1.8527   0.24615    1.1836
## site108               0.5485     1.8233   0.23216    1.2957
## site10R               0.7027     1.4231   0.52323    0.9437
## site11R               0.6027     1.6593   0.42702    0.8506
## site12R               0.4850     2.0617   0.32648    0.7206
## site13R               0.3121     3.2046   0.14301    0.6809
## site14R               0.4701     2.1271   0.32792    0.6740
## site15R               0.7714     1.2964   0.64022    0.9294
## site16R               0.6399     1.5627   0.47052    0.8703
## site17R               0.4257     2.3493   0.22484    0.8058
## site18R               0.4144     2.4131   0.20819    0.8249
## site19R               0.9098     1.0991   0.84089    0.9845
## site20R               0.4297     2.3271   0.27827    0.6636
## site21R               0.4819     2.0750   0.35994    0.6453
## site22R               0.3820     2.6178   0.19649    0.7427
## site23R               0.5049     1.9805   0.34439    0.7403
## site24R               0.5560     1.7986   0.41218    0.7500
## site25R               0.4334     2.3074   0.17680    1.0624
## site26R               0.3822     2.6164   0.19491    0.7495
## site27R               0.4087     2.4467   0.20623    0.8100
## site28R               0.3346     2.9886   0.14266    0.7848
## site29R               0.4089     2.4458   0.23377    0.7151
## site30R               0.2506     3.9903   0.09082    0.6916
## site31R               0.5267     1.8987   0.36498    0.7600
## site32R               0.5274     1.8960   0.35597    0.7815
## site33R               0.4323     2.3133   0.25811    0.7240
## site34R               0.4092     2.4437   0.21675    0.7726
## site35R               0.6659     1.5018   0.54235    0.8175
## site36R               0.2529     3.9538   0.10509    0.6087
## site37R               0.2518     3.9714   0.09291    0.6824
## site38R               0.5068     1.9730   0.37354    0.6877
## site39R               0.5186     1.9283   0.34715    0.7747
## site40R               0.2538     3.9397   0.05712    1.1280
## site41R               0.3969     2.5195   0.22683    0.6945
## site42R               0.4783     2.0908   0.26841    0.8523
## 
## Concordance= 0.68  (se = 0.064 )
## Likelihood ratio test= 2967  on 59 df,   p=<2e-16
## Wald test            = 8.84  on 59 df,   p=1
## Score (logrank) test = 2343  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

So this indicates that there is a HR of 1.78 per IQR increase in NH4 as compared with a HR of 5.41 per 1ug/m3 increase in NH4.

88.4.3 NH4 Variable HR Spline Models

Base model

#First need to make dataframe that only includes patients with a value for event
Allx <- All %>% filter(!is.na(NH4) & !is.na(deadORtx) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(cohort) & !is.na(site) & NH4>0.005 & NH4<2.55)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$NH4, df=3) + Allx$dx_yr + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$NH4, exp(predicted$fit), type="n")
lines(sm.spline(Allx$NH4, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$NH4, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$NH4, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

Complete model

#First need to make dataframe that only includes patients with time_DeathTxCensor
Allx <- All %>% filter(!is.na(NH4) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(deadORtx) & !is.na(age_dx) & !is.na(sex) & !is.na(smokeHx) & !is.na(dich_Race) & !is.na(disadv) & !is.na(site) & NH4>0.005 & NH4<2.55)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$NH4, df=3) + Allx$dx_yr + Allx$age_dx + Allx$sex + Allx$smokeHx + Allx$dich_Race + Allx$disadv + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$NH4, exp(predicted$fit), type="n")
lines(sm.spline(Allx$NH4, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$NH4, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$NH4, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

88.5 BC

88.5.1 BC Continuous Models

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ BC + dx_yr + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC + dx_yr + site, 
##     data = All, id = ID, cluster = cohort)
## 
##   n= 335367, number of events= 6459 
##    (1981 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## BC       0.623396  1.865252  0.075663  0.135620   4.597 4.29e-06 ***
## dx_yr    0.224451  1.251636  0.005420  0.147413   1.523 0.127858    
## site02R  0.431520  1.539595  0.253982  0.135957   3.174 0.001504 ** 
## site03R -0.263787  0.768137  0.263289  0.054179  -4.869 1.12e-06 ***
## site04R -0.313005  0.731247  0.271016  0.049687  -6.300 2.98e-10 ***
## site05R -0.375044  0.687259  0.276191  0.064815  -5.786 7.19e-09 ***
## site06R -0.043820  0.957127  0.267174  0.143003  -0.306 0.759281    
## site07R -0.498761  0.607283  0.245524  0.162795  -3.064 0.002186 ** 
## site09R -0.122208  0.884965  0.269307  0.048677  -2.511 0.012052 *  
## site1   -0.318778  0.727037  0.216915  0.053073  -6.006 1.90e-09 ***
## site101 -0.392850  0.675130  0.220180  0.230311  -1.706 0.088057 .  
## site102 -0.405460  0.666670  0.220117  0.205969  -1.969 0.049005 *  
## site103 -0.147002  0.863292  0.215501  0.150626  -0.976 0.329092    
## site104 -0.344090  0.708865  0.221272  0.159980  -2.151 0.031490 *  
## site105 -0.327090  0.721019  0.217362  0.227804  -1.436 0.151048    
## site106 -0.447346  0.639323  0.218863  0.172019  -2.601 0.009307 ** 
## site107  0.003107  1.003112  0.240679  0.270037   0.012 0.990820    
## site108 -0.060045  0.941722  0.240359  0.420577  -0.143 0.886473    
## site10R -0.183892  0.832025  0.289868  0.156225  -1.177 0.239155    
## site11R -0.273312  0.760855  0.239523  0.048824  -5.598 2.17e-08 ***
## site12R -0.090473  0.913499  0.252312  0.005599 -16.158  < 2e-16 ***
## site13R -0.500289  0.606355  0.240878  0.150536  -3.323 0.000889 ***
## site14R -0.016089  0.984040  0.396074  0.212863  -0.076 0.939751    
## site15R -0.094109  0.910184  0.270979  0.027214  -3.458 0.000544 ***
## site16R  0.025005  1.025320  0.263577  0.099605   0.251 0.801782    
## site17R -0.174267  0.840073  0.275521  0.022825  -7.635 2.26e-14 ***
## site18R -0.519501  0.594817  0.259575  0.253124  -2.052 0.040135 *  
## site19R -0.020643  0.979569  0.282872  0.076743  -0.269 0.787941    
## site20R -0.266995  0.765677  0.287275  0.015747 -16.955  < 2e-16 ***
## site21R -0.518074  0.595667  0.253815  0.049981 -10.365  < 2e-16 ***
## site22R -0.201852  0.817216  0.247915  0.027099  -7.449 9.43e-14 ***
## site23R -0.212928  0.808214  0.248258  0.018201 -11.699  < 2e-16 ***
## site24R  0.020601  1.020814  0.248752  0.094546   0.218 0.827516    
## site25R -0.193998  0.823660  0.250639  0.087121  -2.227 0.025962 *  
## site26R -0.371342  0.689808  0.265489  0.119644  -3.104 0.001911 ** 
## site27R -0.245104  0.782623  0.369424  0.048899  -5.012 5.37e-07 ***
## site28R -0.242614  0.784575  0.266395  0.031409  -7.724 1.12e-14 ***
## site29R -0.372649  0.688907  0.334905  0.099371  -3.750 0.000177 ***
## site30R -0.455788  0.633948  0.255960  0.132196  -3.448 0.000565 ***
## site31R -0.259066  0.771772  0.270718  0.033555  -7.721 1.16e-14 ***
## site32R -0.494954  0.609599  0.268103  0.110189  -4.492 7.06e-06 ***
## site33R -0.313828  0.730645  0.257513  0.072641  -4.320 1.56e-05 ***
## site34R -0.237651  0.788478  0.247689  0.091933  -2.585 0.009736 ** 
## site35R -0.075334  0.927433  0.264290  0.022454  -3.355 0.000793 ***
## site36R -0.404516  0.667299  0.255283  0.036115 -11.201  < 2e-16 ***
## site37R -0.391680  0.675921  0.257259  0.106632  -3.673 0.000240 ***
## site38R -0.369925  0.690786  0.261322  0.034814 -10.626  < 2e-16 ***
## site39R -0.338845  0.712593  0.289895  0.063769  -5.314 1.07e-07 ***
## site40R -0.508129  0.601620  0.274668  0.125490  -4.049 5.14e-05 ***
## site41R -0.422185  0.655613  0.252358  0.058167  -7.258 3.92e-13 ***
## site42R -0.007998  0.992034  0.275752  0.054090  -0.148 0.882454    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## BC         1.8653     0.5361    1.4299    2.4332
## dx_yr      1.2516     0.7990    0.9376    1.6709
## site02R    1.5396     0.6495    1.1794    2.0097
## site03R    0.7681     1.3019    0.6908    0.8542
## site04R    0.7312     1.3675    0.6634    0.8060
## site05R    0.6873     1.4551    0.6053    0.7804
## site06R    0.9571     1.0448    0.7232    1.2668
## site07R    0.6073     1.6467    0.4414    0.8355
## site09R    0.8850     1.1300    0.8044    0.9736
## site1      0.7270     1.3754    0.6552    0.8067
## site101    0.6751     1.4812    0.4299    1.0603
## site102    0.6667     1.5000    0.4452    0.9982
## site103    0.8633     1.1584    0.6426    1.1598
## site104    0.7089     1.4107    0.5181    0.9699
## site105    0.7210     1.3869    0.4614    1.1268
## site106    0.6393     1.5642    0.4563    0.8957
## site107    1.0031     0.9969    0.5909    1.7030
## site108    0.9417     1.0619    0.4130    2.1474
## site10R    0.8320     1.2019    0.6126    1.1301
## site11R    0.7609     1.3143    0.6914    0.8373
## site12R    0.9135     1.0947    0.9035    0.9236
## site13R    0.6064     1.6492    0.4514    0.8144
## site14R    0.9840     1.0162    0.6484    1.4935
## site15R    0.9102     1.0987    0.8629    0.9600
## site16R    1.0253     0.9753    0.8435    1.2464
## site17R    0.8401     1.1904    0.8033    0.8785
## site18R    0.5948     1.6812    0.3622    0.9769
## site19R    0.9796     1.0209    0.8428    1.1386
## site20R    0.7657     1.3060    0.7424    0.7897
## site21R    0.5957     1.6788    0.5401    0.6570
## site22R    0.8172     1.2237    0.7749    0.8618
## site23R    0.8082     1.2373    0.7799    0.8376
## site24R    1.0208     0.9796    0.8481    1.2286
## site25R    0.8237     1.2141    0.6944    0.9770
## site26R    0.6898     1.4497    0.5456    0.8721
## site27R    0.7826     1.2778    0.7111    0.8613
## site28R    0.7846     1.2746    0.7377    0.8344
## site29R    0.6889     1.4516    0.5670    0.8370
## site30R    0.6339     1.5774    0.4892    0.8214
## site31R    0.7718     1.2957    0.7226    0.8242
## site32R    0.6096     1.6404    0.4912    0.7565
## site33R    0.7306     1.3687    0.6337    0.8424
## site34R    0.7885     1.2683    0.6585    0.9442
## site35R    0.9274     1.0782    0.8875    0.9692
## site36R    0.6673     1.4986    0.6217    0.7162
## site37R    0.6759     1.4795    0.5484    0.8330
## site38R    0.6908     1.4476    0.6452    0.7396
## site39R    0.7126     1.4033    0.6289    0.8075
## site40R    0.6016     1.6622    0.4704    0.7694
## site41R    0.6556     1.5253    0.5850    0.7348
## site42R    0.9920     1.0080    0.8922    1.1030
## 
## Concordance= 0.67  (se = 0.065 )
## Likelihood ratio test= 2435  on 51 df,   p=<2e-16
## Wald test            = 99.77  on 51 df,   p=5e-05
## Score (logrank) test = 1986  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ BC + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = All, id = ID, 
##     cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## BC                  0.673552  1.961191  0.077570  0.117165   5.749 8.99e-09 ***
## dx_yr               0.221454  1.247889  0.005575  0.150231   1.474 0.140459    
## age_dx              0.009712  1.009760  0.001176  0.003881   2.503 0.012324 *  
## sexF               -0.139078  0.870160  0.026697  0.059383  -2.342 0.019178 *  
## dich_RaceNon-White -0.021876  0.978361  0.036935  0.025429  -0.860 0.389625    
## smokeHxFormer       0.121988  1.129740  0.032976  0.049859   2.447 0.014420 *  
## smokeHxAlways       0.050602  1.051904  0.074836  0.035797   1.414 0.157478    
## smokeHxUnknown      0.052147  1.053530  0.070625  0.037597   1.387 0.165448    
## smokeHxEver        -0.009495  0.990550  0.049954  0.013453  -0.706 0.480295    
## disadv             -0.040944  0.959883  0.045721  0.060060  -0.682 0.495413    
## site02R             0.386991  1.472543  0.258511  0.127605   3.033 0.002424 ** 
## site03R            -0.277760  0.757479  0.267717  0.046802  -5.935 2.94e-09 ***
## site04R            -0.303759  0.738039  0.275473  0.049839  -6.095 1.10e-09 ***
## site05R            -0.379243  0.684379  0.282509  0.084946  -4.465 8.03e-06 ***
## site06R            -0.101138  0.903808  0.273196  0.181960  -0.556 0.578330    
## site07R            -0.565772  0.567922  0.250240  0.147157  -3.845 0.000121 ***
## site09R            -0.195678  0.822277  0.275732  0.031338  -6.244 4.26e-10 ***
## site1              -0.426565  0.652748  0.224825  0.029250 -14.583  < 2e-16 ***
## site101            -0.440153  0.643938  0.227016  0.202540  -2.173 0.029767 *  
## site102            -0.481747  0.617704  0.227139  0.181988  -2.647 0.008118 ** 
## site103            -0.200948  0.817955  0.222570  0.113322  -1.773 0.076186 .  
## site104            -0.425795  0.653250  0.228170  0.117690  -3.618 0.000297 ***
## site105            -0.412471  0.662012  0.224760  0.193920  -2.127 0.033418 *  
## site106            -0.589617  0.554540  0.226189  0.126919  -4.646 3.39e-06 ***
## site107            -0.083000  0.920351  0.247252  0.235845  -0.352 0.724893    
## site108            -0.152065  0.858932  0.246493  0.381234  -0.399 0.689984    
## site10R            -0.245273  0.782491  0.296604  0.137496  -1.784 0.074447 .  
## site11R            -0.367948  0.692153  0.246178  0.035616 -10.331  < 2e-16 ***
## site12R            -0.148223  0.862239  0.256887  0.021274  -6.967 3.23e-12 ***
## site13R            -0.613442  0.541484  0.245795  0.139230  -4.406 1.05e-05 ***
## site14R            -0.039754  0.961026  0.399130  0.215195  -0.185 0.853438    
## site15R            -0.178205  0.836771  0.275710  0.030542  -5.835 5.39e-09 ***
## site16R             0.015635  1.015758  0.267909  0.100250   0.156 0.876065    
## site17R            -0.160072  0.852082  0.279938  0.016169  -9.900  < 2e-16 ***
## site18R            -0.576027  0.562127  0.264269  0.227828  -2.528 0.011460 *  
## site19R            -0.082257  0.921035  0.287101  0.052684  -1.561 0.118448    
## site20R            -0.312229  0.731814  0.291564  0.015882 -19.660  < 2e-16 ***
## site21R            -0.590003  0.554326  0.258582  0.062101  -9.501  < 2e-16 ***
## site22R            -0.291291  0.747298  0.252947  0.028063 -10.380  < 2e-16 ***
## site23R            -0.256284  0.773922  0.255155  0.016194 -15.826  < 2e-16 ***
## site24R             0.022769  1.023030  0.253591  0.102021   0.223 0.823396    
## site25R            -0.245571  0.782258  0.256956  0.084704  -2.899 0.003742 ** 
## site26R            -0.432789  0.648697  0.270064  0.102716  -4.213 2.52e-05 ***
## site27R            -0.258762  0.772007  0.372569  0.041097  -6.296 3.05e-10 ***
## site28R            -0.193701  0.823904  0.271076  0.011085 -17.474  < 2e-16 ***
## site29R            -0.444298  0.641274  0.339096  0.102215  -4.347 1.38e-05 ***
## site30R            -0.536771  0.584633  0.261318  0.120715  -4.447 8.72e-06 ***
## site31R            -0.283166  0.753394  0.274947  0.024938 -11.355  < 2e-16 ***
## site32R            -0.511520  0.599584  0.278187  0.082449  -6.204 5.50e-10 ***
## site33R            -0.366015  0.693492  0.263026  0.045881  -7.978 1.49e-15 ***
## site34R            -0.304850  0.737234  0.252880  0.080664  -3.779 0.000157 ***
## site35R            -0.152172  0.858840  0.270176  0.014040 -10.839  < 2e-16 ***
## site36R            -0.434888  0.647337  0.260064  0.035142 -12.375  < 2e-16 ***
## site37R            -0.466666  0.627090  0.261932  0.097650  -4.779 1.76e-06 ***
## site38R            -0.416159  0.659575  0.266938  0.024377 -17.072  < 2e-16 ***
## site39R            -0.422743  0.655247  0.294275  0.041289 -10.239  < 2e-16 ***
## site40R            -0.621647  0.537059  0.281199  0.107376  -5.789 7.06e-09 ***
## site41R            -0.462487  0.629716  0.256939  0.047685  -9.699  < 2e-16 ***
## site42R            -0.040619  0.960195  0.280009  0.055601  -0.731 0.465055    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## BC                    1.9612     0.5099    1.5588    2.4675
## dx_yr                 1.2479     0.8014    0.9296    1.6751
## age_dx                1.0098     0.9903    1.0021    1.0175
## sexF                  0.8702     1.1492    0.7746    0.9776
## dich_RaceNon-White    0.9784     1.0221    0.9308    1.0284
## smokeHxFormer         1.1297     0.8852    1.0246    1.2457
## smokeHxAlways         1.0519     0.9507    0.9806    1.1284
## smokeHxUnknown        1.0535     0.9492    0.9787    1.1341
## smokeHxEver           0.9905     1.0095    0.9648    1.0170
## disadv                0.9599     1.0418    0.8533    1.0798
## site02R               1.4725     0.6791    1.1467    1.8910
## site03R               0.7575     1.3202    0.6911    0.8302
## site04R               0.7380     1.3549    0.6694    0.8138
## site05R               0.6844     1.4612    0.5794    0.8084
## site06R               0.9038     1.1064    0.6327    1.2911
## site07R               0.5679     1.7608    0.4256    0.7578
## site09R               0.8223     1.2161    0.7733    0.8744
## site1                 0.6527     1.5320    0.6164    0.6913
## site101               0.6439     1.5529    0.4330    0.9577
## site102               0.6177     1.6189    0.4324    0.8824
## site103               0.8180     1.2226    0.6550    1.0214
## site104               0.6532     1.5308    0.5187    0.8227
## site105               0.6620     1.5105    0.4527    0.9681
## site106               0.5545     1.8033    0.4324    0.7112
## site107               0.9204     1.0865    0.5797    1.4612
## site108               0.8589     1.1642    0.4069    1.8133
## site10R               0.7825     1.2780    0.5976    1.0245
## site11R               0.6922     1.4448    0.6455    0.7422
## site12R               0.8622     1.1598    0.8270    0.8990
## site13R               0.5415     1.8468    0.4122    0.7114
## site14R               0.9610     1.0406    0.6303    1.4652
## site15R               0.8368     1.1951    0.7882    0.8884
## site16R               1.0158     0.9845    0.8346    1.2363
## site17R               0.8521     1.1736    0.8255    0.8795
## site18R               0.5621     1.7790    0.3597    0.8785
## site19R               0.9210     1.0857    0.8307    1.0212
## site20R               0.7318     1.3665    0.7094    0.7550
## site21R               0.5543     1.8040    0.4908    0.6261
## site22R               0.7473     1.3382    0.7073    0.7896
## site23R               0.7739     1.2921    0.7497    0.7989
## site24R               1.0230     0.9775    0.8376    1.2495
## site25R               0.7823     1.2784    0.6626    0.9235
## site26R               0.6487     1.5416    0.5304    0.7934
## site27R               0.7720     1.2953    0.7123    0.8368
## site28R               0.8239     1.2137    0.8062    0.8420
## site29R               0.6413     1.5594    0.5249    0.7835
## site30R               0.5846     1.7105    0.4615    0.7407
## site31R               0.7534     1.3273    0.7175    0.7911
## site32R               0.5996     1.6678    0.5101    0.7047
## site33R               0.6935     1.4420    0.6339    0.7587
## site34R               0.7372     1.3564    0.6294    0.8635
## site35R               0.8588     1.1644    0.8355    0.8828
## site36R               0.6473     1.5448    0.6043    0.6935
## site37R               0.6271     1.5947    0.5179    0.7594
## site38R               0.6596     1.5161    0.6288    0.6919
## site39R               0.6552     1.5261    0.6043    0.7105
## site40R               0.5371     1.8620    0.4351    0.6629
## site41R               0.6297     1.5880    0.5735    0.6914
## site42R               0.9602     1.0415    0.8611    1.0707
## 
## Concordance= 0.674  (se = 0.063 )
## Likelihood ratio test= 2610  on 59 df,   p=<2e-16
## Wald test            = 178  on 59 df,   p=7e-14
## Score (logrank) test = 2154  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

88.5.2 BC Per IQR

summary(All$BC)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##  0.0401  0.4526  0.6276  0.6228  0.7736  2.5448     688

IQR(All$BC, na.rm=T)

## [1] 0.3210303

# Will use the 5yr pre-censoring IQR (0.325731), not this one
All <- All %>% mutate(BC_IQR = BC/0.325731)

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ BC_IQR + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC_IQR + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## BC_IQR              0.219397  1.245325  0.025267  0.038164   5.749 8.99e-09 ***
## dx_yr               0.221454  1.247889  0.005575  0.150231   1.474 0.140459    
## age_dx              0.009712  1.009760  0.001176  0.003881   2.503 0.012324 *  
## sexF               -0.139078  0.870160  0.026697  0.059383  -2.342 0.019178 *  
## dich_RaceNon-White -0.021876  0.978361  0.036935  0.025429  -0.860 0.389625    
## smokeHxFormer       0.121988  1.129740  0.032976  0.049859   2.447 0.014420 *  
## smokeHxAlways       0.050602  1.051904  0.074836  0.035797   1.414 0.157478    
## smokeHxUnknown      0.052147  1.053530  0.070625  0.037597   1.387 0.165448    
## smokeHxEver        -0.009495  0.990550  0.049954  0.013453  -0.706 0.480295    
## disadv             -0.040944  0.959883  0.045721  0.060060  -0.682 0.495413    
## site02R             0.386991  1.472543  0.258511  0.127605   3.033 0.002424 ** 
## site03R            -0.277760  0.757479  0.267717  0.046802  -5.935 2.94e-09 ***
## site04R            -0.303759  0.738039  0.275473  0.049839  -6.095 1.10e-09 ***
## site05R            -0.379243  0.684379  0.282509  0.084946  -4.465 8.03e-06 ***
## site06R            -0.101138  0.903808  0.273196  0.181960  -0.556 0.578330    
## site07R            -0.565772  0.567922  0.250240  0.147157  -3.845 0.000121 ***
## site09R            -0.195678  0.822277  0.275732  0.031338  -6.244 4.26e-10 ***
## site1              -0.426565  0.652748  0.224825  0.029250 -14.583  < 2e-16 ***
## site101            -0.440153  0.643938  0.227016  0.202540  -2.173 0.029767 *  
## site102            -0.481747  0.617704  0.227139  0.181988  -2.647 0.008118 ** 
## site103            -0.200948  0.817955  0.222570  0.113322  -1.773 0.076186 .  
## site104            -0.425795  0.653250  0.228170  0.117690  -3.618 0.000297 ***
## site105            -0.412471  0.662012  0.224760  0.193920  -2.127 0.033418 *  
## site106            -0.589617  0.554540  0.226189  0.126919  -4.646 3.39e-06 ***
## site107            -0.083000  0.920351  0.247252  0.235845  -0.352 0.724893    
## site108            -0.152065  0.858932  0.246493  0.381234  -0.399 0.689984    
## site10R            -0.245273  0.782491  0.296604  0.137496  -1.784 0.074447 .  
## site11R            -0.367948  0.692153  0.246178  0.035616 -10.331  < 2e-16 ***
## site12R            -0.148223  0.862239  0.256887  0.021274  -6.967 3.23e-12 ***
## site13R            -0.613442  0.541484  0.245795  0.139230  -4.406 1.05e-05 ***
## site14R            -0.039754  0.961026  0.399130  0.215195  -0.185 0.853438    
## site15R            -0.178205  0.836771  0.275710  0.030542  -5.835 5.39e-09 ***
## site16R             0.015635  1.015758  0.267909  0.100250   0.156 0.876065    
## site17R            -0.160072  0.852082  0.279938  0.016169  -9.900  < 2e-16 ***
## site18R            -0.576027  0.562127  0.264269  0.227828  -2.528 0.011460 *  
## site19R            -0.082257  0.921035  0.287101  0.052684  -1.561 0.118448    
## site20R            -0.312229  0.731814  0.291564  0.015882 -19.660  < 2e-16 ***
## site21R            -0.590003  0.554326  0.258582  0.062101  -9.501  < 2e-16 ***
## site22R            -0.291291  0.747298  0.252947  0.028063 -10.380  < 2e-16 ***
## site23R            -0.256284  0.773922  0.255155  0.016194 -15.826  < 2e-16 ***
## site24R             0.022769  1.023030  0.253591  0.102021   0.223 0.823396    
## site25R            -0.245571  0.782258  0.256956  0.084704  -2.899 0.003742 ** 
## site26R            -0.432789  0.648697  0.270064  0.102716  -4.213 2.52e-05 ***
## site27R            -0.258762  0.772007  0.372569  0.041097  -6.296 3.05e-10 ***
## site28R            -0.193701  0.823904  0.271076  0.011085 -17.474  < 2e-16 ***
## site29R            -0.444298  0.641274  0.339096  0.102215  -4.347 1.38e-05 ***
## site30R            -0.536771  0.584633  0.261318  0.120715  -4.447 8.72e-06 ***
## site31R            -0.283166  0.753394  0.274947  0.024938 -11.355  < 2e-16 ***
## site32R            -0.511520  0.599584  0.278187  0.082449  -6.204 5.50e-10 ***
## site33R            -0.366015  0.693492  0.263026  0.045881  -7.978 1.49e-15 ***
## site34R            -0.304850  0.737234  0.252880  0.080664  -3.779 0.000157 ***
## site35R            -0.152172  0.858840  0.270176  0.014040 -10.839  < 2e-16 ***
## site36R            -0.434888  0.647337  0.260064  0.035142 -12.375  < 2e-16 ***
## site37R            -0.466666  0.627090  0.261932  0.097650  -4.779 1.76e-06 ***
## site38R            -0.416159  0.659575  0.266938  0.024377 -17.072  < 2e-16 ***
## site39R            -0.422743  0.655247  0.294275  0.041289 -10.239  < 2e-16 ***
## site40R            -0.621647  0.537059  0.281199  0.107376  -5.789 7.06e-09 ***
## site41R            -0.462487  0.629716  0.256939  0.047685  -9.699  < 2e-16 ***
## site42R            -0.040619  0.960195  0.280009  0.055601  -0.731 0.465055    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## BC_IQR                1.2453     0.8030    1.1556    1.3420
## dx_yr                 1.2479     0.8014    0.9296    1.6751
## age_dx                1.0098     0.9903    1.0021    1.0175
## sexF                  0.8702     1.1492    0.7746    0.9776
## dich_RaceNon-White    0.9784     1.0221    0.9308    1.0284
## smokeHxFormer         1.1297     0.8852    1.0246    1.2457
## smokeHxAlways         1.0519     0.9507    0.9806    1.1284
## smokeHxUnknown        1.0535     0.9492    0.9787    1.1341
## smokeHxEver           0.9905     1.0095    0.9648    1.0170
## disadv                0.9599     1.0418    0.8533    1.0798
## site02R               1.4725     0.6791    1.1467    1.8910
## site03R               0.7575     1.3202    0.6911    0.8302
## site04R               0.7380     1.3549    0.6694    0.8138
## site05R               0.6844     1.4612    0.5794    0.8084
## site06R               0.9038     1.1064    0.6327    1.2911
## site07R               0.5679     1.7608    0.4256    0.7578
## site09R               0.8223     1.2161    0.7733    0.8744
## site1                 0.6527     1.5320    0.6164    0.6913
## site101               0.6439     1.5529    0.4330    0.9577
## site102               0.6177     1.6189    0.4324    0.8824
## site103               0.8180     1.2226    0.6550    1.0214
## site104               0.6532     1.5308    0.5187    0.8227
## site105               0.6620     1.5105    0.4527    0.9681
## site106               0.5545     1.8033    0.4324    0.7112
## site107               0.9204     1.0865    0.5797    1.4612
## site108               0.8589     1.1642    0.4069    1.8133
## site10R               0.7825     1.2780    0.5976    1.0245
## site11R               0.6922     1.4448    0.6455    0.7422
## site12R               0.8622     1.1598    0.8270    0.8990
## site13R               0.5415     1.8468    0.4122    0.7114
## site14R               0.9610     1.0406    0.6303    1.4652
## site15R               0.8368     1.1951    0.7882    0.8884
## site16R               1.0158     0.9845    0.8346    1.2363
## site17R               0.8521     1.1736    0.8255    0.8795
## site18R               0.5621     1.7790    0.3597    0.8785
## site19R               0.9210     1.0857    0.8307    1.0212
## site20R               0.7318     1.3665    0.7094    0.7550
## site21R               0.5543     1.8040    0.4908    0.6261
## site22R               0.7473     1.3382    0.7073    0.7896
## site23R               0.7739     1.2921    0.7497    0.7989
## site24R               1.0230     0.9775    0.8376    1.2495
## site25R               0.7823     1.2784    0.6626    0.9235
## site26R               0.6487     1.5416    0.5304    0.7934
## site27R               0.7720     1.2953    0.7123    0.8368
## site28R               0.8239     1.2137    0.8062    0.8420
## site29R               0.6413     1.5594    0.5249    0.7835
## site30R               0.5846     1.7105    0.4615    0.7407
## site31R               0.7534     1.3273    0.7175    0.7911
## site32R               0.5996     1.6678    0.5101    0.7047
## site33R               0.6935     1.4420    0.6339    0.7587
## site34R               0.7372     1.3564    0.6294    0.8635
## site35R               0.8588     1.1644    0.8355    0.8828
## site36R               0.6473     1.5448    0.6043    0.6935
## site37R               0.6271     1.5947    0.5179    0.7594
## site38R               0.6596     1.5161    0.6288    0.6919
## site39R               0.6552     1.5261    0.6043    0.7105
## site40R               0.5371     1.8620    0.4351    0.6629
## site41R               0.6297     1.5880    0.5735    0.6914
## site42R               0.9602     1.0415    0.8611    1.0707
## 
## Concordance= 0.674  (se = 0.063 )
## Likelihood ratio test= 2610  on 59 df,   p=<2e-16
## Wald test            = 178  on 59 df,   p=7e-14
## Score (logrank) test = 2154  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

So this indicates that there is a HR of 1.24 per IQR increase in BC as compared with a HR of 1.96 per 1ug/m3 increase in BC.

88.5.3 BC Variable HR Spline Models

Base model

#First need to make dataframe that only includes patients with a value for event
Allx <- All %>% filter(!is.na(BC) & !is.na(deadORtx) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(cohort) & !is.na(site) & BC<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$BC, df=3) + Allx$dx_yr + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$BC, exp(predicted$fit), type="n")
lines(sm.spline(Allx$BC, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$BC, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$BC, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

Complete model

#First need to make dataframe that only includes patients with time_DeathTxCensor
Allx <- All %>% filter(!is.na(BC) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(deadORtx) & !is.na(age_dx) & !is.na(sex) & !is.na(smokeHx) & !is.na(dich_Race) & !is.na(disadv) & !is.na(site) & BC<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$BC, df=3) + Allx$dx_yr + Allx$age_dx + Allx$sex + Allx$smokeHx + Allx$dich_Race + Allx$disadv + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$BC, exp(predicted$fit), type="n")
lines(sm.spline(Allx$BC, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$BC, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$BC, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

88.6 OM

88.6.1 OM Continuous Models

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ OM + dx_yr + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM + dx_yr + site, 
##     data = All, id = ID, cluster = cohort)
## 
##   n= 335367, number of events= 6459 
##    (1981 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## OM      -0.018766  0.981409  0.012899  0.035398  -0.530 0.596018    
## dx_yr    0.218261  1.243912  0.005452  0.151847   1.437 0.150613    
## site02R  0.483316  1.621443  0.254007  0.140111   3.450 0.000562 ***
## site03R -0.218358  0.803837  0.263214  0.099085  -2.204 0.027542 *  
## site04R -0.189271  0.827563  0.270642  0.087201  -2.171 0.029968 *  
## site05R -0.173216  0.840956  0.275463  0.038227  -4.531 5.86e-06 ***
## site06R  0.104855  1.110549  0.266760  0.118316   0.886 0.375497    
## site07R -0.415717  0.659867  0.245391  0.184951  -2.248 0.024595 *  
## site09R -0.072365  0.930191  0.269252  0.090329  -0.801 0.423056    
## site1   -0.109553  0.896235  0.215607  0.088138  -1.243 0.213882    
## site101 -0.371492  0.689705  0.220191  0.250361  -1.484 0.137855    
## site102 -0.451100  0.636927  0.220048  0.212434  -2.123 0.033713 *  
## site103 -0.161757  0.850648  0.215502  0.152965  -1.057 0.290293    
## site104 -0.333394  0.716488  0.221333  0.157415  -2.118 0.034181 *  
## site105 -0.308440  0.734592  0.217572  0.231860  -1.330 0.183423    
## site106 -0.361471  0.696651  0.218776  0.188760  -1.915 0.055495 .  
## site107  0.014799  1.014909  0.240825  0.273165   0.054 0.956794    
## site108 -0.138058  0.871048  0.240266  0.446123  -0.309 0.756969    
## site10R -0.268206  0.764750  0.290010  0.191852  -1.398 0.162119    
## site11R -0.053798  0.947624  0.238982  0.053304  -1.009 0.312844    
## site12R -0.127241  0.880522  0.252414  0.035453  -3.589 0.000332 ***
## site13R -0.379684  0.684078  0.240514  0.182455  -2.081 0.037436 *  
## site14R  0.113967  1.120715  0.395747  0.161881   0.704 0.481422    
## site15R  0.035208  1.035836  0.270731  0.044529   0.791 0.429133    
## site16R  0.145498  1.156615  0.263200  0.061852   2.352 0.018654 *  
## site17R -0.061862  0.940013  0.275274  0.015896  -3.892 9.96e-05 ***
## site18R -0.428079  0.651760  0.259541  0.268016  -1.597 0.110218    
## site19R -0.105090  0.900243  0.283019  0.098447  -1.067 0.285757    
## site20R -0.224421  0.798979  0.287237  0.022349 -10.042  < 2e-16 ***
## site21R -0.264932  0.767258  0.252680  0.012422 -21.327  < 2e-16 ***
## site22R -0.037611  0.963088  0.247156  0.051551  -0.730 0.465645    
## site23R -0.147370  0.862975  0.248186  0.035698  -4.128 3.66e-05 ***
## site24R  0.009953  1.010003  0.248815  0.059968   0.166 0.868178    
## site25R -0.106802  0.898703  0.250535  0.098460  -1.085 0.278043    
## site26R -0.360775  0.697136  0.265510  0.146371  -2.465 0.013709 *  
## site27R -0.163213  0.849411  0.369311  0.081476  -2.003 0.045156 *  
## site28R -0.040415  0.960391  0.265558  0.032526  -1.243 0.214037    
## site29R -0.341701  0.710561  0.334928  0.131875  -2.591 0.009567 ** 
## site30R -0.276499  0.758435  0.255172  0.174497  -1.585 0.113069    
## site31R -0.197409  0.820855  0.270637  0.066663  -2.961 0.003063 ** 
## site32R -0.422961  0.655104  0.267957  0.139428  -3.034 0.002417 ** 
## site33R -0.294912  0.744597  0.257519  0.096918  -3.043 0.002343 ** 
## site34R -0.222286  0.800687  0.247708  0.120158  -1.850 0.064321 .  
## site35R -0.078901  0.924132  0.264319  0.071229  -1.108 0.267990    
## site36R -0.206354  0.813545  0.254537  0.017977 -11.479  < 2e-16 ***
## site37R -0.240822  0.785981  0.256713  0.142297  -1.692 0.090572 .  
## site38R -0.275536  0.759165  0.261154  0.053601  -5.140 2.74e-07 ***
## site39R -0.260609  0.770583  0.289711  0.093779  -2.779 0.005453 ** 
## site40R  0.085396  1.089148  0.267604  0.149285   0.572 0.567299    
## site41R -0.260113  0.770965  0.251686  0.094473  -2.753 0.005900 ** 
## site42R  0.136753  1.146545  0.275260  0.016870   8.106 5.21e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## OM         0.9814     1.0189    0.9156    1.0519
## dx_yr      1.2439     0.8039    0.9237    1.6751
## site02R    1.6214     0.6167    1.2321    2.1339
## site03R    0.8038     1.2440    0.6620    0.9761
## site04R    0.8276     1.2084    0.6976    0.9818
## site05R    0.8410     1.1891    0.7803    0.9064
## site06R    1.1105     0.9005    0.8807    1.4004
## site07R    0.6599     1.5155    0.4592    0.9482
## site09R    0.9302     1.0750    0.7793    1.1104
## site1      0.8962     1.1158    0.7540    1.0652
## site101    0.6897     1.4499    0.4222    1.1266
## site102    0.6369     1.5700    0.4200    0.9659
## site103    0.8506     1.1756    0.6303    1.1480
## site104    0.7165     1.3957    0.5263    0.9754
## site105    0.7346     1.3613    0.4663    1.1572
## site106    0.6967     1.4354    0.4812    1.0085
## site107    1.0149     0.9853    0.5942    1.7336
## site108    0.8710     1.1480    0.3633    2.0882
## site10R    0.7648     1.3076    0.5251    1.1138
## site11R    0.9476     1.0553    0.8536    1.0520
## site12R    0.8805     1.1357    0.8214    0.9439
## site13R    0.6841     1.4618    0.4784    0.9782
## site14R    1.1207     0.8923    0.8160    1.5392
## site15R    1.0358     0.9654    0.9493    1.1303
## site16R    1.1566     0.8646    1.0246    1.3057
## site17R    0.9400     1.0638    0.9112    0.9698
## site18R    0.6518     1.5343    0.3854    1.1021
## site19R    0.9002     1.1108    0.7423    1.0918
## site20R    0.7990     1.2516    0.7647    0.8348
## site21R    0.7673     1.3033    0.7488    0.7862
## site22R    0.9631     1.0383    0.8705    1.0655
## site23R    0.8630     1.1588    0.8047    0.9255
## site24R    1.0100     0.9901    0.8980    1.1360
## site25R    0.8987     1.1127    0.7410    1.0900
## site26R    0.6971     1.4344    0.5233    0.9288
## site27R    0.8494     1.1773    0.7240    0.9965
## site28R    0.9604     1.0412    0.9011    1.0236
## site29R    0.7106     1.4073    0.5487    0.9201
## site30R    0.7584     1.3185    0.5387    1.0677
## site31R    0.8209     1.2182    0.7203    0.9354
## site32R    0.6551     1.5265    0.4985    0.8610
## site33R    0.7446     1.3430    0.6158    0.9004
## site34R    0.8007     1.2489    0.6327    1.0133
## site35R    0.9241     1.0821    0.8037    1.0626
## site36R    0.8135     1.2292    0.7854    0.8427
## site37R    0.7860     1.2723    0.5947    1.0388
## site38R    0.7592     1.3172    0.6835    0.8433
## site39R    0.7706     1.2977    0.6412    0.9261
## site40R    1.0891     0.9181    0.8129    1.4593
## site41R    0.7710     1.2971    0.6406    0.9278
## site42R    1.1465     0.8722    1.1093    1.1851
## 
## Concordance= 0.673  (se = 0.067 )
## Likelihood ratio test= 2370  on 51 df,   p=<2e-16
## Wald test            = 518.4  on 51 df,   p=<2e-16
## Score (logrank) test = 1907  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ OM + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = All, id = ID, 
##     cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se        z Pr(>|z|)
## OM                 -0.015514  0.984606  0.013218  0.035472   -0.437 0.661852
## dx_yr               0.215136  1.240030  0.005610  0.155057    1.387 0.165302
## age_dx              0.009982  1.010032  0.001177  0.003947    2.529 0.011447
## sexF               -0.142140  0.867500  0.026679  0.059842   -2.375 0.017538
## dich_RaceNon-White  0.008508  1.008544  0.036841  0.028203    0.302 0.762899
## smokeHxFormer       0.102854  1.108329  0.032935  0.062810    1.638 0.101520
## smokeHxAlways       0.046812  1.047925  0.074822  0.020030    2.337 0.019438
## smokeHxUnknown      0.061387  1.063310  0.070527  0.038421    1.598 0.110103
## smokeHxEver        -0.006450  0.993571  0.049895  0.013031   -0.495 0.620640
## disadv              0.012674  1.012755  0.045925  0.024629    0.515 0.606831
## site02R             0.467637  1.596218  0.258448  0.128658    3.635 0.000278
## site03R            -0.201480  0.817520  0.267526  0.089346   -2.255 0.024129
## site04R            -0.137578  0.871467  0.274870  0.087801   -1.567 0.117133
## site05R            -0.143354  0.866448  0.281557  0.067067   -2.137 0.032559
## site06R             0.073315  1.076069  0.272683  0.151487    0.484 0.628411
## site07R            -0.451129  0.636909  0.249995  0.167214   -2.698 0.006977
## site09R            -0.115890  0.890573  0.275521  0.066768   -1.736 0.082614
## site1              -0.164466  0.848347  0.223150  0.057886   -2.841 0.004494
## site101            -0.381233  0.683019  0.227042  0.215548   -1.769 0.076949
## site102            -0.483281  0.616757  0.227195  0.174286   -2.773 0.005556
## site103            -0.192101  0.825223  0.222705  0.111223   -1.727 0.084138
## site104            -0.386610  0.679356  0.228355  0.109416   -3.533 0.000410
## site105            -0.353697  0.702088  0.225163  0.189645   -1.865 0.062175
## site106            -0.455873  0.633894  0.226026  0.133517   -3.414 0.000639
## site107            -0.041749  0.959111  0.247540  0.231878   -0.180 0.857115
## site108            -0.196684  0.821450  0.246534  0.399508   -0.492 0.622496
## site10R            -0.310210  0.733293  0.296755  0.176684   -1.756 0.079134
## site11R            -0.117228  0.889383  0.245473  0.037089   -3.161 0.001574
## site12R            -0.157987  0.853861  0.256945  0.022693   -6.962 3.36e-12
## site13R            -0.452055  0.636319  0.245178  0.163219   -2.770 0.005612
## site14R             0.131940  1.141040  0.398621  0.176470    0.748 0.454663
## site15R            -0.028651  0.971756  0.275362  0.034480   -0.831 0.406011
## site16R             0.172630  1.188427  0.267366  0.071230    2.424 0.015369
## site17R            -0.014834  0.985276  0.279548  0.020565   -0.721 0.470712
## site18R            -0.466637  0.627108  0.264184  0.248757   -1.876 0.060673
## site19R            -0.139784  0.869546  0.287202  0.075887   -1.842 0.065474
## site20R            -0.235590  0.790105  0.291412  0.009749  -24.166  < 2e-16
## site21R            -0.304958  0.737154  0.257247  0.021083  -14.465  < 2e-16
## site22R            -0.083502  0.919889  0.251856  0.034841   -2.397 0.016546
## site23R            -0.170428  0.843303  0.254994  0.012582  -13.546  < 2e-16
## site24R             0.043897  1.044875  0.253560  0.071933    0.610 0.541693
## site25R            -0.129850  0.878227  0.256770  0.091032   -1.426 0.153749
## site26R            -0.411664  0.662547  0.270086  0.135854   -3.030 0.002444
## site27R            -0.142220  0.867430  0.372359  0.073700   -1.930 0.053640
## site28R             0.032543  1.033079  0.270123  0.037966    0.857 0.391348
## site29R            -0.366688  0.693026  0.338916  0.122079   -3.004 0.002667
## site30R            -0.323406  0.723680  0.260298  0.160519   -2.015 0.043930
## site31R            -0.189253  0.827577  0.274758  0.058590   -3.230 0.001237
## site32R            -0.415191  0.660214  0.277947  0.121356   -3.421 0.000623
## site33R            -0.331322  0.717974  0.262978  0.073612   -4.501 6.77e-06
## site34R            -0.257530  0.772959  0.252810  0.104895   -2.455 0.014084
## site35R            -0.126847  0.880868  0.270110  0.045929   -2.762 0.005749
## site36R            -0.203835  0.815597  0.259132  0.001470 -138.682  < 2e-16
## site37R            -0.278190  0.757153  0.261156  0.130070   -2.139 0.032454
## site38R            -0.295580  0.744100  0.266684  0.045663   -6.473 9.60e-11
## site39R            -0.306678  0.735888  0.293860  0.067064   -4.573 4.81e-06
## site40R             0.032720  1.033261  0.273248  0.122009    0.268 0.788562
## site41R            -0.264936  0.767255  0.256055  0.084266   -3.144 0.001666
## site42R             0.141569  1.152079  0.279315  0.028402    4.984 6.21e-07
##                       
## OM                    
## dx_yr                 
## age_dx             *  
## sexF               *  
## dich_RaceNon-White    
## smokeHxFormer         
## smokeHxAlways      *  
## smokeHxUnknown        
## smokeHxEver           
## disadv                
## site02R            ***
## site03R            *  
## site04R               
## site05R            *  
## site06R               
## site07R            ** 
## site09R            .  
## site1              ** 
## site101            .  
## site102            ** 
## site103            .  
## site104            ***
## site105            .  
## site106            ***
## site107               
## site108               
## site10R            .  
## site11R            ** 
## site12R            ***
## site13R            ** 
## site14R               
## site15R               
## site16R            *  
## site17R               
## site18R            .  
## site19R            .  
## site20R            ***
## site21R            ***
## site22R            *  
## site23R            ***
## site24R               
## site25R               
## site26R            ** 
## site27R            .  
## site28R               
## site29R            ** 
## site30R            *  
## site31R            ** 
## site32R            ***
## site33R            ***
## site34R            *  
## site35R            ** 
## site36R            ***
## site37R            *  
## site38R            ***
## site39R            ***
## site40R               
## site41R            ** 
## site42R            ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## OM                    0.9846     1.0156    0.9185    1.0555
## dx_yr                 1.2400     0.8064    0.9151    1.6804
## age_dx                1.0100     0.9901    1.0022    1.0179
## sexF                  0.8675     1.1527    0.7715    0.9755
## dich_RaceNon-White    1.0085     0.9915    0.9543    1.0659
## smokeHxFormer         1.1083     0.9023    0.9800    1.2535
## smokeHxAlways         1.0479     0.9543    1.0076    1.0899
## smokeHxUnknown        1.0633     0.9405    0.9862    1.1465
## smokeHxEver           0.9936     1.0065    0.9685    1.0193
## disadv                1.0128     0.9874    0.9650    1.0628
## site02R               1.5962     0.6265    1.2404    2.0540
## site03R               0.8175     1.2232    0.6862    0.9740
## site04R               0.8715     1.1475    0.7337    1.0351
## site05R               0.8664     1.1541    0.7597    0.9882
## site06R               1.0761     0.9293    0.7996    1.4481
## site07R               0.6369     1.5701    0.4589    0.8839
## site09R               0.8906     1.1229    0.7813    1.0151
## site1                 0.8483     1.1788    0.7574    0.9503
## site101               0.6830     1.4641    0.4477    1.0421
## site102               0.6168     1.6214    0.4383    0.8679
## site103               0.8252     1.2118    0.6636    1.0262
## site104               0.6794     1.4720    0.5482    0.8418
## site105               0.7021     1.4243    0.4841    1.0182
## site106               0.6339     1.5775    0.4879    0.8235
## site107               0.9591     1.0426    0.6088    1.5109
## site108               0.8215     1.2174    0.3754    1.7974
## site10R               0.7333     1.3637    0.5187    1.0367
## site11R               0.8894     1.1244    0.8270    0.9564
## site12R               0.8539     1.1712    0.8167    0.8927
## site13R               0.6363     1.5715    0.4621    0.8762
## site14R               1.1410     0.8764    0.8074    1.6125
## site15R               0.9718     1.0291    0.9083    1.0397
## site16R               1.1884     0.8414    1.0336    1.3665
## site17R               0.9853     1.0149    0.9464    1.0258
## site18R               0.6271     1.5946    0.3851    1.0211
## site19R               0.8695     1.1500    0.7494    1.0090
## site20R               0.7901     1.2657    0.7752    0.8053
## site21R               0.7372     1.3566    0.7073    0.7683
## site22R               0.9199     1.0871    0.8592    0.9849
## site23R               0.8433     1.1858    0.8228    0.8644
## site24R               1.0449     0.9571    0.9075    1.2031
## site25R               0.8782     1.1387    0.7347    1.0498
## site26R               0.6625     1.5093    0.5077    0.8647
## site27R               0.8674     1.1528    0.7508    1.0022
## site28R               1.0331     0.9680    0.9590    1.1129
## site29R               0.6930     1.4429    0.5456    0.8804
## site30R               0.7237     1.3818    0.5283    0.9912
## site31R               0.8276     1.2083    0.7378    0.9283
## site32R               0.6602     1.5147    0.5205    0.8375
## site33R               0.7180     1.3928    0.6215    0.8294
## site34R               0.7730     1.2937    0.6293    0.9494
## site35R               0.8809     1.1352    0.8050    0.9638
## site36R               0.8156     1.2261    0.8133    0.8180
## site37R               0.7572     1.3207    0.5868    0.9770
## site38R               0.7441     1.3439    0.6804    0.8138
## site39R               0.7359     1.3589    0.6452    0.8393
## site40R               1.0333     0.9678    0.8135    1.3124
## site41R               0.7673     1.3033    0.6504    0.9050
## site42R               1.1521     0.8680    1.0897    1.2180
## 
## Concordance= 0.677  (se = 0.064 )
## Likelihood ratio test= 2537  on 59 df,   p=<2e-16
## Wald test            = 446.8  on 59 df,   p=<2e-16
## Score (logrank) test = 2066  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

88.6.2 OM Per IQR

summary(All$OM)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##  0.1262  2.3201  2.9292  3.0532  3.7305 10.8459     688

IQR(All$OM, na.rm=T)

## [1] 1.410373

# Will use the 5yr pre-censoring IQR (1.270977), not this one
All <- All %>% mutate(OM_IQR = OM/1.270977)

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ OM_IQR + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM_IQR + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se        z Pr(>|z|)
## OM_IQR             -0.019718  0.980475  0.016800  0.045084   -0.437 0.661852
## dx_yr               0.215136  1.240030  0.005610  0.155057    1.387 0.165302
## age_dx              0.009982  1.010032  0.001177  0.003947    2.529 0.011447
## sexF               -0.142140  0.867500  0.026679  0.059842   -2.375 0.017538
## dich_RaceNon-White  0.008508  1.008544  0.036841  0.028203    0.302 0.762899
## smokeHxFormer       0.102854  1.108329  0.032935  0.062810    1.638 0.101520
## smokeHxAlways       0.046812  1.047925  0.074822  0.020030    2.337 0.019438
## smokeHxUnknown      0.061387  1.063310  0.070527  0.038421    1.598 0.110103
## smokeHxEver        -0.006450  0.993571  0.049895  0.013031   -0.495 0.620640
## disadv              0.012674  1.012755  0.045925  0.024629    0.515 0.606831
## site02R             0.467637  1.596218  0.258448  0.128658    3.635 0.000278
## site03R            -0.201480  0.817520  0.267526  0.089346   -2.255 0.024129
## site04R            -0.137578  0.871467  0.274870  0.087801   -1.567 0.117133
## site05R            -0.143354  0.866448  0.281557  0.067067   -2.137 0.032559
## site06R             0.073315  1.076069  0.272683  0.151487    0.484 0.628411
## site07R            -0.451129  0.636909  0.249995  0.167214   -2.698 0.006977
## site09R            -0.115890  0.890573  0.275521  0.066768   -1.736 0.082614
## site1              -0.164466  0.848347  0.223150  0.057886   -2.841 0.004494
## site101            -0.381233  0.683019  0.227042  0.215548   -1.769 0.076949
## site102            -0.483281  0.616757  0.227195  0.174286   -2.773 0.005556
## site103            -0.192101  0.825223  0.222705  0.111223   -1.727 0.084138
## site104            -0.386610  0.679356  0.228355  0.109416   -3.533 0.000410
## site105            -0.353697  0.702088  0.225163  0.189645   -1.865 0.062175
## site106            -0.455873  0.633894  0.226026  0.133517   -3.414 0.000639
## site107            -0.041749  0.959111  0.247540  0.231878   -0.180 0.857115
## site108            -0.196684  0.821450  0.246534  0.399508   -0.492 0.622496
## site10R            -0.310210  0.733293  0.296755  0.176684   -1.756 0.079134
## site11R            -0.117228  0.889383  0.245473  0.037089   -3.161 0.001574
## site12R            -0.157987  0.853861  0.256945  0.022693   -6.962 3.36e-12
## site13R            -0.452055  0.636319  0.245178  0.163219   -2.770 0.005612
## site14R             0.131940  1.141040  0.398621  0.176470    0.748 0.454663
## site15R            -0.028651  0.971756  0.275362  0.034480   -0.831 0.406011
## site16R             0.172630  1.188427  0.267366  0.071230    2.424 0.015369
## site17R            -0.014834  0.985276  0.279548  0.020565   -0.721 0.470712
## site18R            -0.466637  0.627108  0.264184  0.248757   -1.876 0.060673
## site19R            -0.139784  0.869546  0.287202  0.075887   -1.842 0.065474
## site20R            -0.235590  0.790105  0.291412  0.009749  -24.166  < 2e-16
## site21R            -0.304958  0.737154  0.257247  0.021083  -14.465  < 2e-16
## site22R            -0.083502  0.919889  0.251856  0.034841   -2.397 0.016546
## site23R            -0.170428  0.843303  0.254994  0.012582  -13.546  < 2e-16
## site24R             0.043897  1.044875  0.253560  0.071933    0.610 0.541693
## site25R            -0.129850  0.878227  0.256770  0.091032   -1.426 0.153749
## site26R            -0.411664  0.662547  0.270086  0.135854   -3.030 0.002444
## site27R            -0.142220  0.867430  0.372359  0.073700   -1.930 0.053640
## site28R             0.032543  1.033079  0.270123  0.037966    0.857 0.391348
## site29R            -0.366688  0.693026  0.338916  0.122079   -3.004 0.002667
## site30R            -0.323406  0.723680  0.260298  0.160519   -2.015 0.043930
## site31R            -0.189253  0.827577  0.274758  0.058590   -3.230 0.001237
## site32R            -0.415191  0.660214  0.277947  0.121356   -3.421 0.000623
## site33R            -0.331322  0.717974  0.262978  0.073612   -4.501 6.77e-06
## site34R            -0.257530  0.772959  0.252810  0.104895   -2.455 0.014084
## site35R            -0.126847  0.880868  0.270110  0.045929   -2.762 0.005749
## site36R            -0.203835  0.815597  0.259132  0.001470 -138.682  < 2e-16
## site37R            -0.278190  0.757153  0.261156  0.130070   -2.139 0.032454
## site38R            -0.295580  0.744100  0.266684  0.045663   -6.473 9.60e-11
## site39R            -0.306678  0.735888  0.293860  0.067064   -4.573 4.81e-06
## site40R             0.032720  1.033261  0.273248  0.122009    0.268 0.788562
## site41R            -0.264936  0.767255  0.256055  0.084266   -3.144 0.001666
## site42R             0.141569  1.152079  0.279315  0.028402    4.984 6.21e-07
##                       
## OM_IQR                
## dx_yr                 
## age_dx             *  
## sexF               *  
## dich_RaceNon-White    
## smokeHxFormer         
## smokeHxAlways      *  
## smokeHxUnknown        
## smokeHxEver           
## disadv                
## site02R            ***
## site03R            *  
## site04R               
## site05R            *  
## site06R               
## site07R            ** 
## site09R            .  
## site1              ** 
## site101            .  
## site102            ** 
## site103            .  
## site104            ***
## site105            .  
## site106            ***
## site107               
## site108               
## site10R            .  
## site11R            ** 
## site12R            ***
## site13R            ** 
## site14R               
## site15R               
## site16R            *  
## site17R               
## site18R            .  
## site19R            .  
## site20R            ***
## site21R            ***
## site22R            *  
## site23R            ***
## site24R               
## site25R               
## site26R            ** 
## site27R            .  
## site28R               
## site29R            ** 
## site30R            *  
## site31R            ** 
## site32R            ***
## site33R            ***
## site34R            *  
## site35R            ** 
## site36R            ***
## site37R            *  
## site38R            ***
## site39R            ***
## site40R               
## site41R            ** 
## site42R            ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## OM_IQR                0.9805     1.0199    0.8976    1.0711
## dx_yr                 1.2400     0.8064    0.9151    1.6804
## age_dx                1.0100     0.9901    1.0022    1.0179
## sexF                  0.8675     1.1527    0.7715    0.9755
## dich_RaceNon-White    1.0085     0.9915    0.9543    1.0659
## smokeHxFormer         1.1083     0.9023    0.9800    1.2535
## smokeHxAlways         1.0479     0.9543    1.0076    1.0899
## smokeHxUnknown        1.0633     0.9405    0.9862    1.1465
## smokeHxEver           0.9936     1.0065    0.9685    1.0193
## disadv                1.0128     0.9874    0.9650    1.0628
## site02R               1.5962     0.6265    1.2404    2.0540
## site03R               0.8175     1.2232    0.6862    0.9740
## site04R               0.8715     1.1475    0.7337    1.0351
## site05R               0.8664     1.1541    0.7597    0.9882
## site06R               1.0761     0.9293    0.7996    1.4481
## site07R               0.6369     1.5701    0.4589    0.8839
## site09R               0.8906     1.1229    0.7813    1.0151
## site1                 0.8483     1.1788    0.7574    0.9503
## site101               0.6830     1.4641    0.4477    1.0421
## site102               0.6168     1.6214    0.4383    0.8679
## site103               0.8252     1.2118    0.6636    1.0262
## site104               0.6794     1.4720    0.5482    0.8418
## site105               0.7021     1.4243    0.4841    1.0182
## site106               0.6339     1.5775    0.4879    0.8235
## site107               0.9591     1.0426    0.6088    1.5109
## site108               0.8215     1.2174    0.3754    1.7974
## site10R               0.7333     1.3637    0.5187    1.0367
## site11R               0.8894     1.1244    0.8270    0.9564
## site12R               0.8539     1.1712    0.8167    0.8927
## site13R               0.6363     1.5715    0.4621    0.8762
## site14R               1.1410     0.8764    0.8074    1.6125
## site15R               0.9718     1.0291    0.9083    1.0397
## site16R               1.1884     0.8414    1.0336    1.3665
## site17R               0.9853     1.0149    0.9464    1.0258
## site18R               0.6271     1.5946    0.3851    1.0211
## site19R               0.8695     1.1500    0.7494    1.0090
## site20R               0.7901     1.2657    0.7752    0.8053
## site21R               0.7372     1.3566    0.7073    0.7683
## site22R               0.9199     1.0871    0.8592    0.9849
## site23R               0.8433     1.1858    0.8228    0.8644
## site24R               1.0449     0.9571    0.9075    1.2031
## site25R               0.8782     1.1387    0.7347    1.0498
## site26R               0.6625     1.5093    0.5077    0.8647
## site27R               0.8674     1.1528    0.7508    1.0022
## site28R               1.0331     0.9680    0.9590    1.1129
## site29R               0.6930     1.4429    0.5456    0.8804
## site30R               0.7237     1.3818    0.5283    0.9912
## site31R               0.8276     1.2083    0.7378    0.9283
## site32R               0.6602     1.5147    0.5205    0.8375
## site33R               0.7180     1.3928    0.6215    0.8294
## site34R               0.7730     1.2937    0.6293    0.9494
## site35R               0.8809     1.1352    0.8050    0.9638
## site36R               0.8156     1.2261    0.8133    0.8180
## site37R               0.7572     1.3207    0.5868    0.9770
## site38R               0.7441     1.3439    0.6804    0.8138
## site39R               0.7359     1.3589    0.6452    0.8393
## site40R               1.0333     0.9678    0.8135    1.3124
## site41R               0.7673     1.3033    0.6504    0.9050
## site42R               1.1521     0.8680    1.0897    1.2180
## 
## Concordance= 0.677  (se = 0.064 )
## Likelihood ratio test= 2537  on 59 df,   p=<2e-16
## Wald test            = 446.8  on 59 df,   p=<2e-16
## Score (logrank) test = 2066  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

So this indicates that there is a HR of 0.98 per IQR increase in OM as compared with a HR of 0.98 per 1ug/m3 increase in OM.

88.6.3 OM Variable HR Spline Models

Base model

#First need to make dataframe that only includes patients with a value for event
Allx <- All %>% filter(!is.na(OM) & !is.na(deadORtx) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(cohort) & !is.na(site) & OM<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$OM, df=3) + Allx$dx_yr + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$OM, exp(predicted$fit), type="n")
lines(sm.spline(Allx$OM, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$OM, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$OM, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

Complete model

#First need to make dataframe that only includes patients with time_DeathTxCensor
Allx <- All %>% filter(!is.na(OM) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(deadORtx) & !is.na(age_dx) & !is.na(sex) & !is.na(smokeHx) & !is.na(dich_Race) & !is.na(disadv) & !is.na(site) & OM<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$OM, df=3) + Allx$dx_yr + Allx$age_dx + Allx$sex + Allx$smokeHx + Allx$dich_Race + Allx$disadv + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$OM, exp(predicted$fit), type="n")
lines(sm.spline(Allx$OM, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$OM, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$OM, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

88.7 SS

88.7.1 SS Continuous Models

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ SS + dx_yr + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS + dx_yr + site, 
##     data = All, id = ID, cluster = cohort)
## 
##   n= 335367, number of events= 6459 
##    (1981 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## SS       0.020497  1.020708  0.094091  0.292279   0.070 0.944092    
## dx_yr    0.218675  1.244426  0.005446  0.153801   1.422 0.155084    
## site02R  0.451455  1.570596  0.277807  0.208896   2.161 0.030684 *  
## site03R -0.215314  0.806289  0.263528  0.131153  -1.642 0.100653    
## site04R -0.187276  0.829215  0.270664  0.093410  -2.005 0.044976 *  
## site05R -0.183586  0.832280  0.275726  0.034778  -5.279 1.30e-07 ***
## site06R  0.098593  1.103617  0.267155  0.052636   1.873 0.061051 .  
## site07R -0.418419  0.658086  0.245434  0.206932  -2.022 0.043175 *  
## site09R -0.072048  0.930486  0.269305  0.067387  -1.069 0.284990    
## site1   -0.114678  0.891653  0.215741  0.127950  -0.896 0.370105    
## site101 -0.370024  0.690718  0.220332  0.272609  -1.357 0.174673    
## site102 -0.445296  0.640634  0.221248  0.279195  -1.595 0.110728    
## site103 -0.163740  0.848963  0.215499  0.158340  -1.034 0.301087    
## site104 -0.340244  0.711596  0.221295  0.178542  -1.906 0.056691 .  
## site105 -0.319591  0.726446  0.217530  0.275479  -1.160 0.245997    
## site106 -0.368383  0.691852  0.218759  0.216263  -1.703 0.088493 .  
## site107  0.005055  1.005068  0.240814  0.313941   0.016 0.987153    
## site108 -0.125552  0.882009  0.241458  0.503898  -0.249 0.803235    
## site10R -0.265710  0.766661  0.298862  0.083472  -3.183 0.001456 ** 
## site11R -0.076462  0.926388  0.238459  0.108904  -0.702 0.482616    
## site12R -0.111015  0.894926  0.252962  0.065656  -1.691 0.090864 .  
## site13R -0.380287  0.683665  0.240591  0.203923  -1.865 0.062202 .  
## site14R  0.117091  1.124222  0.395749  0.178262   0.657 0.511276    
## site15R  0.022114  1.022360  0.271056  0.022713   0.974 0.330260    
## site16R  0.141641  1.152163  0.263384  0.090836   1.559 0.118924    
## site17R -0.063763  0.938227  0.275689  0.069463  -0.918 0.358651    
## site18R -0.441491  0.643077  0.259948  0.236392  -1.868 0.061815 .  
## site19R -0.084947  0.918561  0.282927  0.094557  -0.898 0.368990    
## site20R -0.218766  0.803510  0.287767  0.068424  -3.197 0.001388 ** 
## site21R -0.283005  0.753516  0.252429  0.062441  -4.532 5.83e-06 ***
## site22R -0.040274  0.960527  0.247285  0.029654  -1.358 0.174434    
## site23R -0.150835  0.859989  0.248180  0.047534  -3.173 0.001508 ** 
## site24R  0.022773  1.023034  0.249230  0.032946   0.691 0.489418    
## site25R -0.138316  0.870824  0.270341  0.210071  -0.658 0.510266    
## site26R -0.366130  0.693413  0.269581  0.040022  -9.148  < 2e-16 ***
## site27R -0.158290  0.853602  0.369315  0.081007  -1.954 0.050696 .  
## site28R -0.048549  0.952611  0.265782  0.090530  -0.536 0.591770    
## site29R -0.341095  0.710991  0.336781  0.033922 -10.055  < 2e-16 ***
## site30R -0.279651  0.756048  0.255174  0.189059  -1.479 0.139094    
## site31R -0.191355  0.825840  0.270954  0.098655  -1.940 0.052424 .  
## site32R -0.422518  0.655394  0.267986  0.150555  -2.806 0.005010 ** 
## site33R -0.295941  0.743831  0.258600  0.028789 -10.280  < 2e-16 ***
## site34R -0.215616  0.806044  0.247820  0.133272  -1.618 0.105690    
## site35R -0.075115  0.927637  0.264363  0.038549  -1.949 0.051348 .  
## site36R -0.217525  0.804507  0.254571  0.068009  -3.198 0.001381 ** 
## site37R -0.243247  0.784078  0.256762  0.164639  -1.477 0.139552    
## site38R -0.279114  0.756454  0.261190  0.077336  -3.609 0.000307 ***
## site39R -0.258297  0.772366  0.290266  0.148974  -1.734 0.082948 .  
## site40R  0.033132  1.033687  0.267245  0.149033   0.222 0.824071    
## site41R -0.266855  0.765784  0.251833  0.073843  -3.614 0.000302 ***
## site42R  0.135382  1.144974  0.275270  0.006286  21.536  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## SS         1.0207     0.9797    0.5756    1.8100
## dx_yr      1.2444     0.8036    0.9206    1.6822
## site02R    1.5706     0.6367    1.0429    2.3653
## site03R    0.8063     1.2403    0.6235    1.0426
## site04R    0.8292     1.2060    0.6905    0.9958
## site05R    0.8323     1.2015    0.7774    0.8910
## site06R    1.1036     0.9061    0.9954    1.2236
## site07R    0.6581     1.5196    0.4387    0.9872
## site09R    0.9305     1.0747    0.8154    1.0619
## site1      0.8917     1.1215    0.6939    1.1458
## site101    0.6907     1.4478    0.4048    1.1785
## site102    0.6406     1.5610    0.3706    1.1073
## site103    0.8490     1.1779    0.6225    1.1579
## site104    0.7116     1.4053    0.5015    1.0097
## site105    0.7264     1.3766    0.4234    1.2465
## site106    0.6919     1.4454    0.4528    1.0571
## site107    1.0051     0.9950    0.5432    1.8596
## site108    0.8820     1.1338    0.3285    2.3681
## site10R    0.7667     1.3044    0.6510    0.9029
## site11R    0.9264     1.0795    0.7483    1.1468
## site12R    0.8949     1.1174    0.7869    1.0178
## site13R    0.6837     1.4627    0.4584    1.0196
## site14R    1.1242     0.8895    0.7927    1.5944
## site15R    1.0224     0.9781    0.9778    1.0689
## site16R    1.1522     0.8679    0.9643    1.3767
## site17R    0.9382     1.0658    0.8188    1.0751
## site18R    0.6431     1.5550    0.4046    1.0221
## site19R    0.9186     1.0887    0.7632    1.1056
## site20R    0.8035     1.2445    0.7027    0.9188
## site21R    0.7535     1.3271    0.6667    0.8516
## site22R    0.9605     1.0411    0.9063    1.0180
## site23R    0.8600     1.1628    0.7835    0.9440
## site24R    1.0230     0.9775    0.9591    1.0913
## site25R    0.8708     1.1483    0.5769    1.3144
## site26R    0.6934     1.4421    0.6411    0.7500
## site27R    0.8536     1.1715    0.7283    1.0005
## site28R    0.9526     1.0497    0.7977    1.1376
## site29R    0.7110     1.4065    0.6653    0.7599
## site30R    0.7560     1.3227    0.5219    1.0952
## site31R    0.8258     1.2109    0.6806    1.0020
## site32R    0.6554     1.5258    0.4879    0.8803
## site33R    0.7438     1.3444    0.7030    0.7870
## site34R    0.8060     1.2406    0.6208    1.0466
## site35R    0.9276     1.0780    0.8601    1.0004
## site36R    0.8045     1.2430    0.7041    0.9192
## site37R    0.7841     1.2754    0.5678    1.0827
## site38R    0.7565     1.3220    0.6501    0.8803
## site39R    0.7724     1.2947    0.5768    1.0343
## site40R    1.0337     0.9674    0.7718    1.3843
## site41R    0.7658     1.3059    0.6626    0.8850
## site42R    1.1450     0.8734    1.1310    1.1592
## 
## Concordance= 0.673  (se = 0.067 )
## Likelihood ratio test= 2368  on 51 df,   p=<2e-16
## Wald test            = 43.91  on 51 df,   p=0.7
## Score (logrank) test = 1916  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SS + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = All, id = ID, 
##     cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## SS                 -0.003421  0.996585  0.095956  0.288188  -0.012  0.99053    
## dx_yr               0.215533  1.240523  0.005608  0.156930   1.373  0.16962    
## age_dx              0.009958  1.010007  0.001177  0.003922   2.539  0.01113 *  
## sexF               -0.141662  0.867914  0.026679  0.060851  -2.328  0.01991 *  
## dich_RaceNon-White  0.005131  1.005144  0.036820  0.026930   0.191  0.84890    
## smokeHxFormer       0.105476  1.111239  0.032878  0.053466   1.973  0.04852 *  
## smokeHxAlways       0.048142  1.049320  0.074843  0.027627   1.743  0.08141 .  
## smokeHxUnknown      0.061380  1.063303  0.070543  0.038375   1.599  0.10971    
## smokeHxEver        -0.006733  0.993290  0.049901  0.013133  -0.513  0.60819    
## disadv              0.004896  1.004908  0.045440  0.042582   0.115  0.90846    
## site02R             0.463006  1.588843  0.283019  0.202948   2.281  0.02252 *  
## site03R            -0.204403  0.815133  0.267780  0.125842  -1.624  0.10432    
## site04R            -0.140085  0.869284  0.274869  0.097138  -1.442  0.14926    
## site05R            -0.157114  0.854606  0.281673  0.009075 -17.314  < 2e-16 ***
## site06R             0.063789  1.065867  0.272947  0.090322   0.706  0.48004    
## site07R            -0.457045  0.633152  0.249967  0.188675  -2.422  0.01542 *  
## site09R            -0.116798  0.889765  0.275657  0.042175  -2.769  0.00562 ** 
## site1              -0.174802  0.839623  0.223111  0.097668  -1.790  0.07349 .  
## site101            -0.385609  0.680036  0.227107  0.243254  -1.585  0.11292    
## site102            -0.489002  0.613238  0.228233  0.251062  -1.948  0.05145 .  
## site103            -0.196579  0.821537  0.222664  0.120906  -1.626  0.10398    
## site104            -0.396062  0.672965  0.228221  0.135455  -2.924  0.00346 ** 
## site105            -0.369283  0.691230  0.224872  0.240335  -1.537  0.12441    
## site106            -0.467452  0.626597  0.225840  0.168203  -2.779  0.00545 ** 
## site107            -0.054793  0.946681  0.247379  0.277989  -0.197  0.84374    
## site108            -0.195814  0.822165  0.247665  0.464570  -0.421  0.67339    
## site10R            -0.294130  0.745179  0.306458  0.092343  -3.185  0.00145 ** 
## site11R            -0.138452  0.870705  0.244822  0.091878  -1.507  0.13183    
## site12R            -0.151133  0.859733  0.257411  0.050688  -2.982  0.00287 ** 
## site13R            -0.457266  0.633012  0.245183  0.186741  -2.449  0.01434 *  
## site14R             0.131260  1.140264  0.398650  0.189175   0.694  0.48777    
## site15R            -0.037164  0.963518  0.275815  0.023588  -1.576  0.11513    
## site16R             0.168619  1.183669  0.267615  0.097843   1.723  0.08482 .  
## site17R            -0.021922  0.978316  0.279824  0.065434  -0.335  0.73760    
## site18R            -0.474957  0.621912  0.264707  0.212787  -2.232  0.02561 *  
## site19R            -0.128027  0.879830  0.287125  0.070387  -1.819  0.06893 .  
## site20R            -0.237786  0.788372  0.291829  0.058812  -4.043 5.27e-05 ***
## site21R            -0.323098  0.723903  0.256867  0.047140  -6.854 7.18e-12 ***
## site22R            -0.087242  0.916455  0.252042  0.016525  -5.279 1.30e-07 ***
## site23R            -0.174929  0.839517  0.254969  0.023360  -7.488 6.97e-14 ***
## site24R             0.047989  1.049159  0.253908  0.046753   1.026  0.30469    
## site25R            -0.136346  0.872541  0.277087  0.207245  -0.658  0.51060    
## site26R            -0.406370  0.666064  0.274476  0.041580  -9.773  < 2e-16 ***
## site27R            -0.141574  0.867991  0.372362  0.075396  -1.878  0.06042 .  
## site28R             0.022352  1.022603  0.270262  0.088201   0.253  0.79995    
## site29R            -0.363318  0.695365  0.341137  0.019319 -18.806  < 2e-16 ***
## site30R            -0.328410  0.720067  0.260267  0.174927  -1.877  0.06046 .  
## site31R            -0.189745  0.827170  0.275005  0.092108  -2.060  0.03940 *  
## site32R            -0.417309  0.658817  0.277950  0.130516  -3.197  0.00139 ** 
## site33R            -0.328079  0.720306  0.264237  0.027493 -11.933  < 2e-16 ***
## site34R            -0.256751  0.773561  0.252885  0.118765  -2.162  0.03063 *  
## site35R            -0.125270  0.882258  0.270239  0.018260  -6.860 6.87e-12 ***
## site36R            -0.217007  0.804924  0.259036  0.053442  -4.061 4.89e-05 ***
## site37R            -0.284064  0.752718  0.261139  0.152718  -1.860  0.06288 .  
## site38R            -0.301446  0.739748  0.266659  0.067173  -4.488 7.20e-06 ***
## site39R            -0.311576  0.732292  0.294282  0.121934  -2.555  0.01061 *  
## site40R            -0.006397  0.993623  0.272799  0.122780  -0.052  0.95845    
## site41R            -0.270591  0.762929  0.256254  0.062240  -4.348 1.38e-05 ***
## site42R             0.137083  1.146923  0.279290  0.017544   7.814 5.55e-15 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SS                    0.9966     1.0034    0.5665    1.7532
## dx_yr                 1.2405     0.8061    0.9121    1.6873
## age_dx                1.0100     0.9901    1.0023    1.0178
## sexF                  0.8679     1.1522    0.7703    0.9779
## dich_RaceNon-White    1.0051     0.9949    0.9535    1.0596
## smokeHxFormer         1.1112     0.8999    1.0007    1.2340
## smokeHxAlways         1.0493     0.9530    0.9940    1.1077
## smokeHxUnknown        1.0633     0.9405    0.9863    1.1464
## smokeHxEver           0.9933     1.0068    0.9680    1.0192
## disadv                1.0049     0.9951    0.9244    1.0924
## site02R               1.5888     0.6294    1.0674    2.3650
## site03R               0.8151     1.2268    0.6370    1.0431
## site04R               0.8693     1.1504    0.7186    1.0516
## site05R               0.8546     1.1701    0.8395    0.8699
## site06R               1.0659     0.9382    0.8929    1.2723
## site07R               0.6332     1.5794    0.4374    0.9164
## site09R               0.8898     1.1239    0.8192    0.9664
## site1                 0.8396     1.1910    0.6933    1.0168
## site101               0.6800     1.4705    0.4222    1.0954
## site102               0.6132     1.6307    0.3749    1.0031
## site103               0.8215     1.2172    0.6482    1.0412
## site104               0.6730     1.4860    0.5161    0.8776
## site105               0.6912     1.4467    0.4316    1.1071
## site106               0.6266     1.5959    0.4506    0.8713
## site107               0.9467     1.0563    0.5490    1.6324
## site108               0.8222     1.2163    0.3308    2.0436
## site10R               0.7452     1.3420    0.6218    0.8930
## site11R               0.8707     1.1485    0.7272    1.0425
## site12R               0.8597     1.1632    0.7784    0.9495
## site13R               0.6330     1.5797    0.4390    0.9128
## site14R               1.1403     0.8770    0.7870    1.6521
## site15R               0.9635     1.0379    0.9200    1.0091
## site16R               1.1837     0.8448    0.9771    1.4339
## site17R               0.9783     1.0222    0.8606    1.1122
## site18R               0.6219     1.6079    0.4098    0.9437
## site19R               0.8798     1.1366    0.7665    1.0100
## site20R               0.7884     1.2684    0.7025    0.8847
## site21R               0.7239     1.3814    0.6600    0.7940
## site22R               0.9165     1.0912    0.8872    0.9466
## site23R               0.8395     1.1912    0.8019    0.8788
## site24R               1.0492     0.9531    0.9573    1.1498
## site25R               0.8725     1.1461    0.5813    1.3098
## site26R               0.6661     1.5014    0.6139    0.7226
## site27R               0.8680     1.1521    0.7488    1.0062
## site28R               1.0226     0.9779    0.8603    1.2156
## site29R               0.6954     1.4381    0.6695    0.7222
## site30R               0.7201     1.3888    0.5111    1.0145
## site31R               0.8272     1.2089    0.6905    0.9908
## site32R               0.6588     1.5179    0.5101    0.8509
## site33R               0.7203     1.3883    0.6825    0.7602
## site34R               0.7736     1.2927    0.6129    0.9763
## site35R               0.8823     1.1335    0.8512    0.9144
## site36R               0.8049     1.2424    0.7249    0.8938
## site37R               0.7527     1.3285    0.5580    1.0154
## site38R               0.7397     1.3518    0.6485    0.8438
## site39R               0.7323     1.3656    0.5766    0.9300
## site40R               0.9936     1.0064    0.7811    1.2640
## site41R               0.7629     1.3107    0.6753    0.8619
## site42R               1.1469     0.8719    1.1082    1.1870
## 
## Concordance= 0.676  (se = 0.064 )
## Likelihood ratio test= 2535  on 59 df,   p=<2e-16
## Wald test            = 137.4  on 59 df,   p=3e-08
## Score (logrank) test = 2074  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

88.7.2 SS Per IQR

summary(All$SS)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##  0.0000  0.1591  0.2311  0.2601  0.2982  2.7091     688

IQR(All$SS, na.rm=T)

## [1] 0.1391352

# Will use the 5yr pre-censoring IQR (0.135), not this one
All <- All %>% mutate(SS_IQR = SS/0.135)

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SS_IQR + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS_IQR + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)  robust se       z Pr(>|z|)
## SS_IQR             -0.0004618  0.9995383  0.0129541  0.0389053  -0.012  0.99053
## dx_yr               0.2155331  1.2405230  0.0056076  0.1569298   1.373  0.16962
## age_dx              0.0099576  1.0100074  0.0011774  0.0039223   2.539  0.01113
## sexF               -0.1416622  0.8679144  0.0266794  0.0608514  -2.328  0.01991
## dich_RaceNon-White  0.0051309  1.0051441  0.0368199  0.0269303   0.191  0.84890
## smokeHxFormer       0.1054756  1.1112390  0.0328780  0.0534662   1.973  0.04852
## smokeHxAlways       0.0481420  1.0493197  0.0748429  0.0276273   1.743  0.08141
## smokeHxUnknown      0.0613803  1.0633032  0.0705428  0.0383751   1.599  0.10971
## smokeHxEver        -0.0067328  0.9932898  0.0499011  0.0131331  -0.513  0.60819
## disadv              0.0048959  1.0049079  0.0454403  0.0425824   0.115  0.90846
## site02R             0.4630059  1.5888427  0.2830187  0.2029480   2.281  0.02252
## site03R            -0.2044034  0.8151335  0.2677797  0.1258423  -1.624  0.10432
## site04R            -0.1400855  0.8692839  0.2748688  0.0971376  -1.442  0.14926
## site05R            -0.1571143  0.8546064  0.2816726  0.0090746 -17.314  < 2e-16
## site06R             0.0637889  1.0658674  0.2729475  0.0903224   0.706  0.48004
## site07R            -0.4570452  0.6331517  0.2499672  0.1886746  -2.422  0.01542
## site09R            -0.1167980  0.8897649  0.2756567  0.0421748  -2.769  0.00562
## site1              -0.1748022  0.8396231  0.2231107  0.0976683  -1.790  0.07349
## site101            -0.3856090  0.6800363  0.2271067  0.2432541  -1.585  0.11292
## site102            -0.4890022  0.6132380  0.2282333  0.2510621  -1.948  0.05145
## site103            -0.1965789  0.8215365  0.2226635  0.1209063  -1.626  0.10398
## site104            -0.3960621  0.6729649  0.2282211  0.1354551  -2.924  0.00346
## site105            -0.3692825  0.6912301  0.2248719  0.2403348  -1.537  0.12441
## site106            -0.4674516  0.6265970  0.2258402  0.1682031  -2.779  0.00545
## site107            -0.0547933  0.9466808  0.2473794  0.2779894  -0.197  0.84374
## site108            -0.1958140  0.8221651  0.2476652  0.4645696  -0.421  0.67339
## site10R            -0.2941303  0.7451794  0.3064579  0.0923428  -3.185  0.00145
## site11R            -0.1384518  0.8707053  0.2448223  0.0918779  -1.507  0.13183
## site12R            -0.1511328  0.8597335  0.2574114  0.0506885  -2.982  0.00287
## site13R            -0.4572663  0.6330117  0.2451828  0.1867408  -2.449  0.01434
## site14R             0.1312602  1.1402645  0.3986498  0.1891748   0.694  0.48777
## site15R            -0.0371644  0.9635177  0.2758154  0.0235884  -1.576  0.11513
## site16R             0.1686187  1.1836688  0.2676154  0.0978428   1.723  0.08482
## site17R            -0.0219222  0.9783164  0.2798241  0.0654340  -0.335  0.73760
## site18R            -0.4749573  0.6219116  0.2647071  0.2127873  -2.232  0.02561
## site19R            -0.1280266  0.8798300  0.2871248  0.0703868  -1.819  0.06893
## site20R            -0.2377858  0.7883716  0.2918292  0.0588122  -4.043 5.27e-05
## site21R            -0.3230979  0.7239030  0.2568667  0.0471397  -6.854 7.18e-12
## site22R            -0.0872419  0.9164554  0.2520417  0.0165252  -5.279 1.30e-07
## site23R            -0.1749289  0.8395167  0.2549691  0.0233601  -7.488 6.97e-14
## site24R             0.0479893  1.0491594  0.2539084  0.0467534   1.026  0.30469
## site25R            -0.1363458  0.8725409  0.2770867  0.2072452  -0.658  0.51060
## site26R            -0.4063702  0.6660636  0.2744757  0.0415799  -9.773  < 2e-16
## site27R            -0.1415738  0.8679911  0.3723620  0.0753958  -1.878  0.06042
## site28R             0.0223515  1.0226032  0.2702615  0.0882014   0.253  0.79995
## site29R            -0.3633181  0.6953652  0.3411375  0.0193192 -18.806  < 2e-16
## site30R            -0.3284104  0.7200674  0.2602675  0.1749272  -1.877  0.06046
## site31R            -0.1897447  0.8271703  0.2750048  0.0921078  -2.060  0.03940
## site32R            -0.4173091  0.6588172  0.2779502  0.1305162  -3.197  0.00139
## site33R            -0.3280791  0.7203061  0.2642367  0.0274929 -11.933  < 2e-16
## site34R            -0.2567508  0.7735609  0.2528849  0.1187646  -2.162  0.03063
## site35R            -0.1252702  0.8822585  0.2702391  0.0182603  -6.860 6.87e-12
## site36R            -0.2170070  0.8049243  0.2590360  0.0534418  -4.061 4.89e-05
## site37R            -0.2840644  0.7527182  0.2611385  0.1527177  -1.860  0.06288
## site38R            -0.3014462  0.7397476  0.2666595  0.0671730  -4.488 7.20e-06
## site39R            -0.3115761  0.7322919  0.2942818  0.1219339  -2.555  0.01061
## site40R            -0.0063972  0.9936232  0.2727986  0.1227797  -0.052  0.95845
## site41R            -0.2705906  0.7629288  0.2562540  0.0622400  -4.348 1.38e-05
## site42R             0.1370830  1.1469233  0.2792899  0.0175439   7.814 5.55e-15
##                       
## SS_IQR                
## dx_yr                 
## age_dx             *  
## sexF               *  
## dich_RaceNon-White    
## smokeHxFormer      *  
## smokeHxAlways      .  
## smokeHxUnknown        
## smokeHxEver           
## disadv                
## site02R            *  
## site03R               
## site04R               
## site05R            ***
## site06R               
## site07R            *  
## site09R            ** 
## site1              .  
## site101               
## site102            .  
## site103               
## site104            ** 
## site105               
## site106            ** 
## site107               
## site108               
## site10R            ** 
## site11R               
## site12R            ** 
## site13R            *  
## site14R               
## site15R               
## site16R            .  
## site17R               
## site18R            *  
## site19R            .  
## site20R            ***
## site21R            ***
## site22R            ***
## site23R            ***
## site24R               
## site25R               
## site26R            ***
## site27R            .  
## site28R               
## site29R            ***
## site30R            .  
## site31R            *  
## site32R            ** 
## site33R            ***
## site34R            *  
## site35R            ***
## site36R            ***
## site37R            .  
## site38R            ***
## site39R            *  
## site40R               
## site41R            ***
## site42R            ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SS_IQR                0.9995     1.0005    0.9262    1.0787
## dx_yr                 1.2405     0.8061    0.9121    1.6873
## age_dx                1.0100     0.9901    1.0023    1.0178
## sexF                  0.8679     1.1522    0.7703    0.9779
## dich_RaceNon-White    1.0051     0.9949    0.9535    1.0596
## smokeHxFormer         1.1112     0.8999    1.0007    1.2340
## smokeHxAlways         1.0493     0.9530    0.9940    1.1077
## smokeHxUnknown        1.0633     0.9405    0.9863    1.1464
## smokeHxEver           0.9933     1.0068    0.9680    1.0192
## disadv                1.0049     0.9951    0.9244    1.0924
## site02R               1.5888     0.6294    1.0674    2.3650
## site03R               0.8151     1.2268    0.6370    1.0431
## site04R               0.8693     1.1504    0.7186    1.0516
## site05R               0.8546     1.1701    0.8395    0.8699
## site06R               1.0659     0.9382    0.8929    1.2723
## site07R               0.6332     1.5794    0.4374    0.9164
## site09R               0.8898     1.1239    0.8192    0.9664
## site1                 0.8396     1.1910    0.6933    1.0168
## site101               0.6800     1.4705    0.4222    1.0954
## site102               0.6132     1.6307    0.3749    1.0031
## site103               0.8215     1.2172    0.6482    1.0412
## site104               0.6730     1.4860    0.5161    0.8776
## site105               0.6912     1.4467    0.4316    1.1071
## site106               0.6266     1.5959    0.4506    0.8713
## site107               0.9467     1.0563    0.5490    1.6324
## site108               0.8222     1.2163    0.3308    2.0436
## site10R               0.7452     1.3420    0.6218    0.8930
## site11R               0.8707     1.1485    0.7272    1.0425
## site12R               0.8597     1.1632    0.7784    0.9495
## site13R               0.6330     1.5797    0.4390    0.9128
## site14R               1.1403     0.8770    0.7870    1.6521
## site15R               0.9635     1.0379    0.9200    1.0091
## site16R               1.1837     0.8448    0.9771    1.4339
## site17R               0.9783     1.0222    0.8606    1.1122
## site18R               0.6219     1.6079    0.4098    0.9437
## site19R               0.8798     1.1366    0.7665    1.0100
## site20R               0.7884     1.2684    0.7025    0.8847
## site21R               0.7239     1.3814    0.6600    0.7940
## site22R               0.9165     1.0912    0.8872    0.9466
## site23R               0.8395     1.1912    0.8019    0.8788
## site24R               1.0492     0.9531    0.9573    1.1498
## site25R               0.8725     1.1461    0.5813    1.3098
## site26R               0.6661     1.5014    0.6139    0.7226
## site27R               0.8680     1.1521    0.7488    1.0062
## site28R               1.0226     0.9779    0.8603    1.2156
## site29R               0.6954     1.4381    0.6695    0.7222
## site30R               0.7201     1.3888    0.5111    1.0145
## site31R               0.8272     1.2089    0.6905    0.9908
## site32R               0.6588     1.5179    0.5101    0.8509
## site33R               0.7203     1.3883    0.6825    0.7602
## site34R               0.7736     1.2927    0.6129    0.9763
## site35R               0.8823     1.1335    0.8512    0.9144
## site36R               0.8049     1.2424    0.7249    0.8938
## site37R               0.7527     1.3285    0.5580    1.0154
## site38R               0.7397     1.3518    0.6485    0.8438
## site39R               0.7323     1.3656    0.5766    0.9300
## site40R               0.9936     1.0064    0.7811    1.2640
## site41R               0.7629     1.3107    0.6753    0.8619
## site42R               1.1469     0.8719    1.1082    1.1870
## 
## Concordance= 0.676  (se = 0.064 )
## Likelihood ratio test= 2535  on 59 df,   p=<2e-16
## Wald test            = 137.4  on 59 df,   p=3e-08
## Score (logrank) test = 2074  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

So this indicates that there is a HR of 1.00 per IQR increase in SS as compared with a HR of 1.00 per 1ug/m3 increase in SS.

88.7.3 SS Variable HR Spline Models

Base model

#First need to make dataframe that only includes patients with a value for event
Allx <- All %>% filter(!is.na(SS) & !is.na(deadORtx) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(cohort) & !is.na(site) & SS>0)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$SS, df=3) + Allx$dx_yr + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$SS, exp(predicted$fit), type="n")
lines(sm.spline(Allx$SS, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$SS, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$SS, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

Complete model

#First need to make dataframe that only includes patients with time_DeathTxCensor
Allx <- All %>% filter(!is.na(SS) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(deadORtx) & !is.na(age_dx) & !is.na(sex) & !is.na(smokeHx) & !is.na(dich_Race) & !is.na(disadv) & !is.na(site) & SS>0.01 & SS<2.5)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$SS, df=3) + Allx$dx_yr + Allx$age_dx + Allx$sex + Allx$smokeHx + Allx$dich_Race + Allx$disadv + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$SS, exp(predicted$fit), type="n")
lines(sm.spline(Allx$SS, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$SS, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$SS, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

88.8 Soil

88.8.1 Soil Continuous Models

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ Soil + dx_yr + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil + dx_yr + 
##     site, data = All, id = ID, cluster = cohort)
## 
##   n= 335367, number of events= 6459 
##    (1981 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se      z Pr(>|z|)    
## Soil     0.162021  1.175885  0.090275  0.142038  1.141 0.254001    
## dx_yr    0.219065  1.244912  0.005434  0.152456  1.437 0.150746    
## site02R  0.449024  1.566782  0.254384  0.166042  2.704 0.006845 ** 
## site03R -0.318254  0.727418  0.269226  0.157099 -2.026 0.042783 *  
## site04R -0.234561  0.790918  0.271865  0.110504 -2.123 0.033784 *  
## site05R -0.230171  0.794398  0.276343  0.042655 -5.396 6.81e-08 ***
## site06R  0.049520  1.050767  0.267875  0.094601  0.523 0.600649    
## site07R -0.470435  0.624730  0.247021  0.217856 -2.159 0.030820 *  
## site09R -0.071298  0.931184  0.269248  0.081035 -0.880 0.378942    
## site1   -0.149861  0.860828  0.216341  0.106373 -1.409 0.158886    
## site101 -0.399850  0.670421  0.220742  0.259600 -1.540 0.123498    
## site102 -0.465011  0.628128  0.220193  0.213163 -2.181 0.029148 *  
## site103 -0.158763  0.853199  0.215515  0.149930 -1.059 0.289638    
## site104 -0.338132  0.713102  0.221277  0.165751 -2.040 0.041351 *  
## site105 -0.333346  0.716522  0.217481  0.253939 -1.313 0.189284    
## site106 -0.415554  0.659974  0.220205  0.221803 -1.874 0.060996 .  
## site107 -0.008599  0.991438  0.240780  0.293832 -0.029 0.976654    
## site108 -0.152102  0.858901  0.240485  0.437164 -0.348 0.727893    
## site10R -0.366689  0.693025  0.296976  0.217642 -1.685 0.092022 .  
## site11R -0.160036  0.852113  0.242792  0.144944 -1.104 0.269540    
## site12R -0.138565  0.870606  0.252615  0.031196 -4.442 8.92e-06 ***
## site13R -0.396941  0.672373  0.240651  0.189769 -2.092 0.036465 *  
## site14R  0.112572  1.119153  0.395754  0.170823  0.659 0.509897    
## site15R -0.015617  0.984504  0.271588  0.085044 -0.184 0.854302    
## site16R  0.125776  1.134029  0.263387  0.055685  2.259 0.023902 *  
## site17R -0.130192  0.877927  0.277471  0.072117 -1.805 0.071029 .  
## site18R -0.534479  0.585975  0.264932  0.331554 -1.612 0.106953    
## site19R -0.280423  0.755464  0.302758  0.211408 -1.326 0.184690    
## site20R -0.292978  0.746038  0.289887  0.075641 -3.873 0.000107 ***
## site21R -0.350298  0.704478  0.254919  0.089044 -3.934 8.35e-05 ***
## site22R -0.055063  0.946425  0.247326  0.062793 -0.877 0.380542    
## site23R -0.295273  0.744328  0.260787  0.150088 -1.967 0.049145 *  
## site24R -0.003148  0.996857  0.249047  0.075521 -0.042 0.966754    
## site25R -0.152284  0.858744  0.251272  0.133641 -1.140 0.254492    
## site26R -0.515656  0.597109  0.280025  0.237707 -2.169 0.030060 *  
## site27R -0.166537  0.846592  0.369323  0.073851 -2.255 0.024131 *  
## site28R -0.087355  0.916352  0.266191  0.072301 -1.208 0.226969    
## site29R -0.336509  0.714260  0.334886  0.115652 -2.910 0.003618 ** 
## site30R -0.338999  0.712483  0.257223  0.209088 -1.621 0.104948    
## site31R -0.202786  0.816453  0.270668  0.061086 -3.320 0.000901 ***
## site32R -0.693415  0.499866  0.308038  0.328561 -2.110 0.034819 *  
## site33R -0.450384  0.637383  0.272457  0.198453 -2.269 0.023240 *  
## site34R -0.224777  0.798694  0.247718  0.110804 -2.029 0.042499 *  
## site35R -0.061659  0.940203  0.264384  0.048986 -1.259 0.208137    
## site36R -0.264599  0.767514  0.255582  0.070835 -3.735 0.000187 ***
## site37R -0.256630  0.773655  0.256783  0.151622 -1.693 0.090538 .  
## site38R -0.300128  0.740723  0.261358  0.070549 -4.254 2.10e-05 ***
## site39R -0.351213  0.703834  0.293961  0.149639 -2.347 0.018922 *  
## site40R -0.088977  0.914867  0.275196  0.305118 -0.292 0.770581    
## site41R -0.296620  0.743327  0.252285  0.118843 -2.496 0.012564 *  
## site42R  0.119857  1.127336  0.275377  0.018090  6.626 3.46e-11 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## Soil       1.1759     0.8504    0.8901    1.5533
## dx_yr      1.2449     0.8033    0.9234    1.6785
## site02R    1.5668     0.6383    1.1315    2.1694
## site03R    0.7274     1.3747    0.5346    0.9897
## site04R    0.7909     1.2644    0.6369    0.9822
## site05R    0.7944     1.2588    0.7307    0.8637
## site06R    1.0508     0.9517    0.8729    1.2648
## site07R    0.6247     1.6007    0.4076    0.9575
## site09R    0.9312     1.0739    0.7944    1.0915
## site1      0.8608     1.1617    0.6988    1.0604
## site101    0.6704     1.4916    0.4031    1.1151
## site102    0.6281     1.5920    0.4136    0.9539
## site103    0.8532     1.1721    0.6360    1.1446
## site104    0.7131     1.4023    0.5153    0.9868
## site105    0.7165     1.3956    0.4356    1.1786
## site106    0.6600     1.5152    0.4273    1.0194
## site107    0.9914     1.0086    0.5574    1.7635
## site108    0.8589     1.1643    0.3646    2.0233
## site10R    0.6930     1.4429    0.4524    1.0617
## site11R    0.8521     1.1736    0.6414    1.1321
## site12R    0.8706     1.1486    0.8190    0.9255
## site13R    0.6724     1.4873    0.4635    0.9753
## site14R    1.1192     0.8935    0.8007    1.5642
## site15R    0.9845     1.0157    0.8334    1.1631
## site16R    1.1340     0.8818    1.0168    1.2648
## site17R    0.8779     1.1390    0.7622    1.0112
## site18R    0.5860     1.7066    0.3060    1.1223
## site19R    0.7555     1.3237    0.4992    1.1433
## site20R    0.7460     1.3404    0.6432    0.8653
## site21R    0.7045     1.4195    0.5917    0.8388
## site22R    0.9464     1.0566    0.8368    1.0704
## site23R    0.7443     1.3435    0.5546    0.9989
## site24R    0.9969     1.0032    0.8597    1.1559
## site25R    0.8587     1.1645    0.6609    1.1159
## site26R    0.5971     1.6747    0.3747    0.9515
## site27R    0.8466     1.1812    0.7325    0.9784
## site28R    0.9164     1.0913    0.7953    1.0559
## site29R    0.7143     1.4001    0.5694    0.8960
## site30R    0.7125     1.4035    0.4729    1.0734
## site31R    0.8165     1.2248    0.7243    0.9203
## site32R    0.4999     2.0005    0.2625    0.9518
## site33R    0.6374     1.5689    0.4320    0.9404
## site34R    0.7987     1.2520    0.6428    0.9924
## site35R    0.9402     1.0636    0.8541    1.0349
## site36R    0.7675     1.3029    0.6680    0.8818
## site37R    0.7737     1.2926    0.5748    1.0414
## site38R    0.7407     1.3500    0.6451    0.8506
## site39R    0.7038     1.4208    0.5249    0.9437
## site40R    0.9149     1.0931    0.5031    1.6637
## site41R    0.7433     1.3453    0.5889    0.9383
## site42R    1.1273     0.8870    1.0881    1.1680
## 
## Concordance= 0.673  (se = 0.067 )
## Likelihood ratio test= 2371  on 51 df,   p=<2e-16
## Wald test            = 2.33  on 51 df,   p=1
## Score (logrank) test = 1914  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ Soil + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se      z Pr(>|z|)    
## Soil                0.189992  1.209240  0.091974  0.128283  1.481 0.138595    
## dx_yr               0.215875  1.240947  0.005595  0.155559  1.388 0.165219    
## age_dx              0.009946  1.009996  0.001177  0.003952  2.517 0.011852 *  
## sexF               -0.140764  0.868694  0.026685  0.059926 -2.349 0.018826 *  
## dich_RaceNon-White  0.000342  1.000342  0.036806  0.031263  0.011 0.991272    
## smokeHxFormer       0.105491  1.111257  0.032871  0.054122  1.949 0.051279 .  
## smokeHxAlways       0.046502  1.047600  0.074818  0.023691  1.963 0.049669 *  
## smokeHxUnknown      0.061163  1.063072  0.070571  0.038529  1.587 0.112412    
## smokeHxEver        -0.005830  0.994187  0.049908  0.012666 -0.460 0.645336    
## disadv             -0.005678  0.994338  0.045691  0.050672 -0.112 0.910782    
## site02R             0.425809  1.530828  0.258835  0.161365  2.639 0.008320 ** 
## site03R            -0.323553  0.723574  0.273919  0.145616 -2.222 0.026286 *  
## site04R            -0.196804  0.821351  0.276221  0.112149 -1.755 0.079286 .  
## site05R            -0.207465  0.812641  0.282408  0.030762 -6.744 1.54e-11 ***
## site06R             0.012329  1.012406  0.273721  0.121759  0.101 0.919343    
## site07R            -0.517441  0.596044  0.251677  0.207570 -2.493 0.012672 *  
## site09R            -0.118962  0.887842  0.275527  0.064253 -1.851 0.064104 .  
## site1              -0.214379  0.807042  0.223820  0.081312 -2.637 0.008377 ** 
## site101            -0.419047  0.657673  0.227591  0.234302 -1.788 0.073697 .  
## site102            -0.507681  0.601890  0.227356  0.190181 -2.669 0.007597 ** 
## site103            -0.188759  0.827986  0.222695  0.114958 -1.642 0.100594    
## site104            -0.392000  0.675704  0.228220  0.125617 -3.121 0.001805 ** 
## site105            -0.384221  0.680981  0.224898  0.223539 -1.719 0.085649 .  
## site106            -0.523245  0.592595  0.227437  0.180959 -2.892 0.003834 ** 
## site107            -0.068284  0.933995  0.247382  0.261610 -0.261 0.794080    
## site108            -0.220487  0.802128  0.246807  0.402155 -0.548 0.583511    
## site10R            -0.435042  0.647237  0.303988  0.206323 -2.109 0.034983 *  
## site11R            -0.235570  0.790120  0.249244  0.131754 -1.788 0.073784 .  
## site12R            -0.180155  0.835141  0.257266  0.028125 -6.405 1.50e-10 ***
## site13R            -0.477612  0.620263  0.245339  0.180840 -2.641 0.008264 ** 
## site14R             0.123004  1.130888  0.398649  0.173919  0.707 0.479413    
## site15R            -0.085151  0.918374  0.276178  0.078174 -1.089 0.276042    
## site16R             0.145402  1.156505  0.267562  0.052338  2.778 0.005467 ** 
## site17R            -0.094852  0.909508  0.281737  0.051023 -1.859 0.063027 .  
## site18R            -0.587680  0.555615  0.269543  0.313891 -1.872 0.061173 .  
## site19R            -0.357769  0.699234  0.308076  0.191777 -1.866 0.062104 .  
## site20R            -0.321823  0.724826  0.294265  0.061409 -5.241 1.60e-07 ***
## site21R            -0.399744  0.670492  0.259446  0.080336 -4.976 6.50e-07 ***
## site22R            -0.109729  0.896077  0.252068  0.056668 -1.936 0.052827 .  
## site23R            -0.342496  0.709996  0.267545  0.126374 -2.710 0.006725 ** 
## site24R             0.020056  1.020259  0.253908  0.073195  0.274 0.784077    
## site25R            -0.183472  0.832376  0.257508  0.133436 -1.375 0.169139    
## site26R            -0.592786  0.552785  0.284480  0.225334 -2.631 0.008521 ** 
## site27R            -0.151609  0.859324  0.372391  0.071394 -2.124 0.033708 *  
## site28R            -0.017576  0.982578  0.270680  0.060223 -0.292 0.770410    
## site29R            -0.373397  0.688392  0.338953  0.120703 -3.094 0.001978 ** 
## site30R            -0.398005  0.671659  0.262414  0.201572 -1.975 0.048324 *  
## site31R            -0.200675  0.818179  0.274812  0.058286 -3.443 0.000575 ***
## site32R            -0.736278  0.478893  0.318833  0.298080 -2.470 0.013509 *  
## site33R            -0.513666  0.598298  0.277777  0.173827 -2.955 0.003126 ** 
## site34R            -0.267274  0.765463  0.252866  0.103909 -2.572 0.010105 *  
## site35R            -0.113942  0.892310  0.270178  0.030563 -3.728 0.000193 ***
## site36R            -0.269182  0.764005  0.260131  0.056030 -4.804 1.55e-06 ***
## site37R            -0.299866  0.740918  0.261221  0.146134 -2.052 0.040170 *  
## site38R            -0.325017  0.722515  0.266878  0.065056 -4.996 5.86e-07 ***
## site39R            -0.417580  0.658639  0.298395  0.132631 -3.148 0.001641 ** 
## site40R            -0.157565  0.854221  0.280690  0.285008 -0.553 0.580369    
## site41R            -0.309606  0.733736  0.256683  0.113596 -2.725 0.006421 ** 
## site42R             0.118096  1.125352  0.279434  0.012698  9.301  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## Soil                  1.2092     0.8270    0.9404    1.5549
## dx_yr                 1.2409     0.8058    0.9148    1.6833
## age_dx                1.0100     0.9901    1.0022    1.0178
## sexF                  0.8687     1.1512    0.7724    0.9770
## dich_RaceNon-White    1.0003     0.9997    0.9409    1.0636
## smokeHxFormer         1.1113     0.8999    0.9994    1.2356
## smokeHxAlways         1.0476     0.9546    1.0001    1.0974
## smokeHxUnknown        1.0631     0.9407    0.9857    1.1465
## smokeHxEver           0.9942     1.0058    0.9698    1.0192
## disadv                0.9943     1.0057    0.9003    1.0982
## site02R               1.5308     0.6532    1.1158    2.1003
## site03R               0.7236     1.3820    0.5439    0.9626
## site04R               0.8214     1.2175    0.6593    1.0233
## site05R               0.8126     1.2306    0.7651    0.8631
## site06R               1.0124     0.9877    0.7975    1.2853
## site07R               0.5960     1.6777    0.3968    0.8953
## site09R               0.8878     1.1263    0.7828    1.0070
## site1                 0.8070     1.2391    0.6882    0.9465
## site101               0.6577     1.5205    0.4155    1.0410
## site102               0.6019     1.6614    0.4146    0.8738
## site103               0.8280     1.2077    0.6610    1.0372
## site104               0.6757     1.4799    0.5282    0.8643
## site105               0.6810     1.4685    0.4394    1.0554
## site106               0.5926     1.6875    0.4156    0.8449
## site107               0.9340     1.0707    0.5593    1.5597
## site108               0.8021     1.2467    0.3647    1.7642
## site10R               0.6472     1.5450    0.4320    0.9698
## site11R               0.7901     1.2656    0.6103    1.0229
## site12R               0.8351     1.1974    0.7903    0.8825
## site13R               0.6203     1.6122    0.4352    0.8841
## site14R               1.1309     0.8843    0.8042    1.5902
## site15R               0.9184     1.0889    0.7879    1.0704
## site16R               1.1565     0.8647    1.0438    1.2814
## site17R               0.9095     1.0995    0.8230    1.0052
## site18R               0.5556     1.7998    0.3003    1.0279
## site19R               0.6992     1.4301    0.4802    1.0183
## site20R               0.7248     1.3796    0.6426    0.8175
## site21R               0.6705     1.4914    0.5728    0.7848
## site22R               0.8961     1.1160    0.8019    1.0013
## site23R               0.7100     1.4085    0.5542    0.9095
## site24R               1.0203     0.9801    0.8839    1.1776
## site25R               0.8324     1.2014    0.6408    1.0812
## site26R               0.5528     1.8090    0.3554    0.8597
## site27R               0.8593     1.1637    0.7471    0.9884
## site28R               0.9826     1.0177    0.8732    1.1057
## site29R               0.6884     1.4527    0.5434    0.8721
## site30R               0.6717     1.4889    0.4524    0.9971
## site31R               0.8182     1.2222    0.7299    0.9172
## site32R               0.4789     2.0881    0.2670    0.8589
## site33R               0.5983     1.6714    0.4256    0.8412
## site34R               0.7655     1.3064    0.6244    0.9384
## site35R               0.8923     1.1207    0.8404    0.9474
## site36R               0.7640     1.3089    0.6845    0.8527
## site37R               0.7409     1.3497    0.5564    0.9866
## site38R               0.7225     1.3841    0.6360    0.8208
## site39R               0.6586     1.5183    0.5079    0.8542
## site40R               0.8542     1.1707    0.4886    1.4934
## site41R               0.7337     1.3629    0.5873    0.9167
## site42R               1.1254     0.8886    1.0977    1.1537
## 
## Concordance= 0.676  (se = 0.064 )
## Likelihood ratio test= 2540  on 59 df,   p=<2e-16
## Wald test            = 2.63  on 59 df,   p=1
## Score (logrank) test = 2075  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

88.8.2 Soil Per IQR

summary(All$Soil)

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
##  0.0117  0.2843  0.4002  0.4437  0.5253  2.8544     688

IQR(All$Soil, na.rm=T)

## [1] 0.2410509

# Will use the 5yr pre-censoring IQR (0.2614493), not this one
All <- All %>% mutate(Soil_IQR = Soil/0.2614493)

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ Soil_IQR + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil_IQR + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All, 
##     id = ID, cluster = cohort)
## 
##   n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se      z Pr(>|z|)    
## Soil_IQR            0.049673  1.050928  0.024046  0.033539  1.481 0.138595    
## dx_yr               0.215875  1.240947  0.005595  0.155559  1.388 0.165219    
## age_dx              0.009946  1.009996  0.001177  0.003952  2.517 0.011852 *  
## sexF               -0.140764  0.868694  0.026685  0.059926 -2.349 0.018826 *  
## dich_RaceNon-White  0.000342  1.000342  0.036806  0.031263  0.011 0.991272    
## smokeHxFormer       0.105491  1.111257  0.032871  0.054122  1.949 0.051279 .  
## smokeHxAlways       0.046502  1.047600  0.074818  0.023691  1.963 0.049669 *  
## smokeHxUnknown      0.061163  1.063072  0.070571  0.038529  1.587 0.112412    
## smokeHxEver        -0.005830  0.994187  0.049908  0.012666 -0.460 0.645336    
## disadv             -0.005678  0.994338  0.045691  0.050672 -0.112 0.910782    
## site02R             0.425809  1.530828  0.258835  0.161365  2.639 0.008320 ** 
## site03R            -0.323553  0.723574  0.273919  0.145616 -2.222 0.026286 *  
## site04R            -0.196804  0.821351  0.276221  0.112149 -1.755 0.079286 .  
## site05R            -0.207465  0.812641  0.282408  0.030762 -6.744 1.54e-11 ***
## site06R             0.012329  1.012406  0.273721  0.121759  0.101 0.919343    
## site07R            -0.517441  0.596044  0.251677  0.207570 -2.493 0.012672 *  
## site09R            -0.118962  0.887842  0.275527  0.064253 -1.851 0.064104 .  
## site1              -0.214379  0.807042  0.223820  0.081312 -2.637 0.008377 ** 
## site101            -0.419047  0.657673  0.227591  0.234302 -1.788 0.073697 .  
## site102            -0.507681  0.601890  0.227356  0.190181 -2.669 0.007597 ** 
## site103            -0.188759  0.827986  0.222695  0.114958 -1.642 0.100594    
## site104            -0.392000  0.675704  0.228220  0.125617 -3.121 0.001805 ** 
## site105            -0.384221  0.680981  0.224898  0.223539 -1.719 0.085649 .  
## site106            -0.523245  0.592595  0.227437  0.180959 -2.892 0.003834 ** 
## site107            -0.068284  0.933995  0.247382  0.261610 -0.261 0.794080    
## site108            -0.220487  0.802128  0.246807  0.402155 -0.548 0.583511    
## site10R            -0.435042  0.647237  0.303988  0.206323 -2.109 0.034983 *  
## site11R            -0.235570  0.790120  0.249244  0.131754 -1.788 0.073784 .  
## site12R            -0.180155  0.835141  0.257266  0.028125 -6.405 1.50e-10 ***
## site13R            -0.477612  0.620263  0.245339  0.180840 -2.641 0.008264 ** 
## site14R             0.123004  1.130888  0.398649  0.173919  0.707 0.479413    
## site15R            -0.085151  0.918374  0.276178  0.078174 -1.089 0.276042    
## site16R             0.145402  1.156505  0.267562  0.052338  2.778 0.005467 ** 
## site17R            -0.094852  0.909508  0.281737  0.051023 -1.859 0.063027 .  
## site18R            -0.587680  0.555615  0.269543  0.313891 -1.872 0.061173 .  
## site19R            -0.357769  0.699234  0.308076  0.191777 -1.866 0.062104 .  
## site20R            -0.321823  0.724826  0.294265  0.061409 -5.241 1.60e-07 ***
## site21R            -0.399744  0.670492  0.259446  0.080336 -4.976 6.50e-07 ***
## site22R            -0.109729  0.896077  0.252068  0.056668 -1.936 0.052827 .  
## site23R            -0.342496  0.709996  0.267545  0.126374 -2.710 0.006725 ** 
## site24R             0.020056  1.020259  0.253908  0.073195  0.274 0.784077    
## site25R            -0.183472  0.832376  0.257508  0.133436 -1.375 0.169139    
## site26R            -0.592786  0.552785  0.284480  0.225334 -2.631 0.008521 ** 
## site27R            -0.151609  0.859324  0.372391  0.071394 -2.124 0.033708 *  
## site28R            -0.017576  0.982578  0.270680  0.060223 -0.292 0.770410    
## site29R            -0.373397  0.688392  0.338953  0.120703 -3.094 0.001978 ** 
## site30R            -0.398005  0.671659  0.262414  0.201572 -1.975 0.048324 *  
## site31R            -0.200675  0.818179  0.274812  0.058286 -3.443 0.000575 ***
## site32R            -0.736278  0.478893  0.318833  0.298080 -2.470 0.013509 *  
## site33R            -0.513666  0.598298  0.277777  0.173827 -2.955 0.003126 ** 
## site34R            -0.267274  0.765463  0.252866  0.103909 -2.572 0.010105 *  
## site35R            -0.113942  0.892310  0.270178  0.030563 -3.728 0.000193 ***
## site36R            -0.269182  0.764005  0.260131  0.056030 -4.804 1.55e-06 ***
## site37R            -0.299866  0.740918  0.261221  0.146134 -2.052 0.040170 *  
## site38R            -0.325017  0.722515  0.266878  0.065056 -4.996 5.86e-07 ***
## site39R            -0.417580  0.658639  0.298395  0.132631 -3.148 0.001641 ** 
## site40R            -0.157565  0.854221  0.280690  0.285008 -0.553 0.580369    
## site41R            -0.309606  0.733736  0.256683  0.113596 -2.725 0.006421 ** 
## site42R             0.118096  1.125352  0.279434  0.012698  9.301  < 2e-16 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## Soil_IQR              1.0509     0.9515    0.9841    1.1223
## dx_yr                 1.2409     0.8058    0.9148    1.6833
## age_dx                1.0100     0.9901    1.0022    1.0178
## sexF                  0.8687     1.1512    0.7724    0.9770
## dich_RaceNon-White    1.0003     0.9997    0.9409    1.0636
## smokeHxFormer         1.1113     0.8999    0.9994    1.2356
## smokeHxAlways         1.0476     0.9546    1.0001    1.0974
## smokeHxUnknown        1.0631     0.9407    0.9857    1.1465
## smokeHxEver           0.9942     1.0058    0.9698    1.0192
## disadv                0.9943     1.0057    0.9003    1.0982
## site02R               1.5308     0.6532    1.1158    2.1003
## site03R               0.7236     1.3820    0.5439    0.9626
## site04R               0.8214     1.2175    0.6593    1.0233
## site05R               0.8126     1.2306    0.7651    0.8631
## site06R               1.0124     0.9877    0.7975    1.2853
## site07R               0.5960     1.6777    0.3968    0.8953
## site09R               0.8878     1.1263    0.7828    1.0070
## site1                 0.8070     1.2391    0.6882    0.9465
## site101               0.6577     1.5205    0.4155    1.0410
## site102               0.6019     1.6614    0.4146    0.8738
## site103               0.8280     1.2077    0.6610    1.0372
## site104               0.6757     1.4799    0.5282    0.8643
## site105               0.6810     1.4685    0.4394    1.0554
## site106               0.5926     1.6875    0.4156    0.8449
## site107               0.9340     1.0707    0.5593    1.5597
## site108               0.8021     1.2467    0.3647    1.7642
## site10R               0.6472     1.5450    0.4320    0.9698
## site11R               0.7901     1.2656    0.6103    1.0229
## site12R               0.8351     1.1974    0.7903    0.8825
## site13R               0.6203     1.6122    0.4352    0.8841
## site14R               1.1309     0.8843    0.8042    1.5902
## site15R               0.9184     1.0889    0.7879    1.0704
## site16R               1.1565     0.8647    1.0438    1.2814
## site17R               0.9095     1.0995    0.8230    1.0052
## site18R               0.5556     1.7998    0.3003    1.0279
## site19R               0.6992     1.4301    0.4802    1.0183
## site20R               0.7248     1.3796    0.6426    0.8175
## site21R               0.6705     1.4914    0.5728    0.7848
## site22R               0.8961     1.1160    0.8019    1.0013
## site23R               0.7100     1.4085    0.5542    0.9095
## site24R               1.0203     0.9801    0.8839    1.1776
## site25R               0.8324     1.2014    0.6408    1.0812
## site26R               0.5528     1.8090    0.3554    0.8597
## site27R               0.8593     1.1637    0.7471    0.9884
## site28R               0.9826     1.0177    0.8732    1.1057
## site29R               0.6884     1.4527    0.5434    0.8721
## site30R               0.6717     1.4889    0.4524    0.9971
## site31R               0.8182     1.2222    0.7299    0.9172
## site32R               0.4789     2.0881    0.2670    0.8589
## site33R               0.5983     1.6714    0.4256    0.8412
## site34R               0.7655     1.3064    0.6244    0.9384
## site35R               0.8923     1.1207    0.8404    0.9474
## site36R               0.7640     1.3089    0.6845    0.8527
## site37R               0.7409     1.3497    0.5564    0.9866
## site38R               0.7225     1.3841    0.6360    0.8208
## site39R               0.6586     1.5183    0.5079    0.8542
## site40R               0.8542     1.1707    0.4886    1.4934
## site41R               0.7337     1.3629    0.5873    0.9167
## site42R               1.1254     0.8886    1.0977    1.1537
## 
## Concordance= 0.676  (se = 0.064 )
## Likelihood ratio test= 2540  on 59 df,   p=<2e-16
## Wald test            = 2.63  on 59 df,   p=1
## Score (logrank) test = 2075  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

So this indicates that there is a HR of 1.05 per IQR increase in Soil as compared with a HR of 1.21 per 1ug/m3 increase in Soil.

88.8.3 Soil Variable HR Spline Models

Base model

#First need to make dataframe that only includes patients with a value for event
Allx <- All %>% filter(!is.na(Soil) & !is.na(deadORtx) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(cohort) & !is.na(site) & Soil<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$Soil, df=3) + Allx$dx_yr + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$Soil, exp(predicted$fit), type="n")
lines(sm.spline(Allx$Soil, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$Soil, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$Soil, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

Complete model

#First need to make dataframe that only includes patients with time_DeathTxCensor
Allx <- All %>% filter(!is.na(Soil) & !is.na(time_DeathTxCensor) & !is.na(dx_yr) & !is.na(deadORtx) & !is.na(age_dx) & !is.na(sex) & !is.na(smokeHx) & !is.na(dich_Race) & !is.na(disadv) & !is.na(site) & Soil<20)

#Then make survival function
surv1 <- Surv(Allx$start, Allx$end, Allx$event==1)
fit1 <- coxph(surv1 ~ pspline(Allx$Soil, df=3) + Allx$dx_yr + Allx$age_dx + Allx$sex + Allx$smokeHx + Allx$dich_Race + Allx$disadv + cluster(Allx$cohort) + Allx$site)
predicted <- predict(fit1, type="terms", se.fit=T, terms=1)

#Then plot
plot(Allx$Soil, exp(predicted$fit), type="n")
lines(sm.spline(Allx$Soil, exp(predicted$fit)), col = "red" , lty = 1 )
lines(sm.spline(Allx$Soil, exp(predicted$fit + 1.96 * predicted$se)), col = "orange" , lty = 2 )
lines(sm.spline(Allx$Soil, exp(predicted$fit - 1.96 * predicted$se)), col = "orange" , lty = 2 )

89 Quantile-Based G-Computation Multi-Pollutant Survival Analysis - No Bootstrapping

89.1 Simmons Alone

Multi-pollutant model without other covariates aside from dx_yr and site

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr, data=Simm)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr, data = Simm)
## 
##           coef exp(coef) se(coef)      z        p
## SO4    0.47543   1.60871  0.16919  2.810  0.00495
## NH4   -1.34488   0.26057  0.53322 -2.522  0.01166
## NO3    0.07989   1.08317  0.25204  0.317  0.75127
## BC     1.95006   7.02908  0.40813  4.778 1.77e-06
## OM     0.10051   1.10573  0.10046  1.000  0.31707
## SS     2.27672   9.74468  0.32152  7.081 1.43e-12
## Soil  -2.51412   0.08093  0.31892 -7.883 3.19e-15
## dx_yr  0.06206   1.06402  0.01949  3.184  0.00145
## 
## Likelihood ratio test=333.5  on 8 df, p=< 2.2e-16
## n= 88140, number of events= 1373 
##    (1367 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit1 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=Simm[,c(Xnm, 'start', 'end', 'event', 'dx_yr')], q=4)
qc.survfit1

## Scaled effect size (positive direction, sum of positive coefficients = 0.64)
##   NH4    SS    OM    BC 
## 0.323 0.293 0.242 0.142 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.441)
##   NO3  Soil   SO4 
## 0.648 0.224 0.129 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##      Estimate Std. Error Lower CI Upper CI Z value Pr(>|z|)
## psi1 0.198873   0.071873 0.058004  0.33974   2.767 0.005658

#Lastly the HR is reported through the following
exp(qc.survfit1$coef)

##     psi1 
## 1.220027

exp(qc.survfit1$ci)

## [1] 1.059719 1.404585

So the HR of the overall model is 1.22 (95% CI 1.06-1.40) and that is primarily driven by NH4, SS, OM, and BC

Now to plot the findings

plot(qc.survfit1)

Complete multi-pollutant model + dx_yr

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + smokeHx + dich_Race + disadv + dx_yr, data=Simm)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + 
##     smokeHx + dich_Race + disadv + dx_yr, data = Simm)
## 
##                         coef exp(coef)  se(coef)      z        p
## SO4                 0.522906  1.686923  0.178132  2.935  0.00333
## NH4                -1.550522  0.212137  0.566039 -2.739  0.00616
## NO3                 0.141085  1.151523  0.267959  0.527  0.59853
## BC                  2.047748  7.750424  0.430917  4.752 2.01e-06
## OM                  0.129216  1.137936  0.109650  1.178  0.23862
## SS                  2.108554  8.236326  0.343357  6.141 8.20e-10
## Soil               -2.574076  0.076224  0.329523 -7.812 5.65e-15
## age_dx              0.018155  1.018321  0.002535  7.160 8.04e-13
## sexF               -0.333389  0.716491  0.059329 -5.619 1.92e-08
## smokeHxFormer       0.147130  1.158505  0.067921  2.166  0.03030
## smokeHxAlways      -0.345321  0.707993  0.177209 -1.949  0.05134
## smokeHxUnknown     -0.095368  0.909038  0.082127 -1.161  0.24555
## dich_RaceNon-White  0.027279  1.027654  0.087549  0.312  0.75536
## disadv              0.408380  1.504379  0.097342  4.195 2.72e-05
## dx_yr               0.047453  1.048597  0.020304  2.337  0.01943
## 
## Likelihood ratio test=465.4  on 15 df, p=< 2.2e-16
## n= 84895, number of events= 1332 
##    (4612 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit2 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=Simm[,c(Xnm, 'age_dx', 'sex', 'smokeHx', 'dich_Race', 'disadv', 'start', 'end', 'event', 'dx_yr')], q=4)
qc.survfit2

## Scaled effect size (positive direction, sum of positive coefficients = 0.632)
##   NH4    OM    SS    BC 
## 0.298 0.288 0.269 0.145 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.455)
##   NO3  Soil   SO4 
## 0.644 0.236 0.121 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##      Estimate Std. Error Lower CI Upper CI Z value Pr(>|z|)
## psi1 0.176884   0.073614 0.032602  0.32117  2.4028  0.01627

#Lastly the HR is reported through the following
exp(qc.survfit2$coef)

##     psi1 
## 1.193492

exp(qc.survfit2$ci)

## [1] 1.033140 1.378733

So the HR of the overall model is 1.19 (95% CI 1.03-1.38) and that is primarily driven by NH4, OM, SS, and BC

Now to plot the findings

plot(qc.survfit2)

89.2 PFF Alone

Multi-pollutant model without other covariates aside from dx_yr and site

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site, data=PFF)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site, 
##     data = PFF)
## 
##              coef exp(coef)  se(coef)      z        p
## SO4      0.031253  1.031746  0.189895  0.165  0.86928
## NH4      0.927134  2.527256  0.457183  2.028  0.04257
## NO3     -0.499240  0.606992  0.167536 -2.980  0.00288
## BC      -0.417934  0.658406  0.368037 -1.136  0.25613
## OM       0.178600  1.195542  0.076881  2.323  0.02018
## SS       0.121441  1.129123  0.123984  0.979  0.32734
## Soil    -0.193370  0.824177  0.187490 -1.031  0.30237
## dx_yr    0.145370  1.156467  0.019092  7.614 2.65e-14
## site02R  0.653902  1.923030  0.297720  2.196  0.02807
## site03R  0.067826  1.070179  0.288888  0.235  0.81438
## site04R  0.065264  1.067441  0.332954  0.196  0.84460
## site05R -0.434478  0.647603  0.339364 -1.280  0.20045
## site06R -0.158683  0.853267  0.316983 -0.501  0.61665
## site07R -0.215107  0.806455  0.297833 -0.722  0.47015
## site08R        NA        NA  0.000000     NA       NA
## site09R -0.085011  0.918503  0.287742 -0.295  0.76766
## site10R  0.012452  1.012530  0.361899  0.034  0.97255
## site11R -0.200321  0.818468  0.301064 -0.665  0.50581
## site12R  0.127611  1.136111  0.290929  0.439  0.66093
## site13R -0.328377  0.720091  0.287513 -1.142  0.25340
## site14R -0.062098  0.939791  0.426642 -0.146  0.88428
## site15R -0.095477  0.908939  0.315948 -0.302  0.76251
## site16R  0.074250  1.077076  0.290663  0.255  0.79837
## site17R -0.007258  0.992768  0.329755 -0.022  0.98244
## site18R -0.351325  0.703755  0.315308 -1.114  0.26518
## site19R  0.290196  1.336690  0.376919  0.770  0.44135
## site20R -0.068691  0.933615  0.334863 -0.205  0.83747
## site21R -0.404267  0.667466  0.313879 -1.288  0.19776
## site22R -0.022023  0.978218  0.293081 -0.075  0.94010
## site23R -0.108428  0.897243  0.327640 -0.331  0.74069
## site24R  0.187573  1.206318  0.292015  0.642  0.52065
## site25R  0.113415  1.120097  0.294927  0.385  0.70057
## site26R -0.363457  0.695269  0.365980 -0.993  0.32066
## site27R -0.158854  0.853121  0.403232 -0.394  0.69362
## site28R -0.179976  0.835290  0.333617 -0.539  0.58956
## site29R -0.315595  0.729355  0.359737 -0.877  0.38033
## site30R  0.075660  1.078596  0.325401  0.233  0.81614
## site31R -0.310188  0.733309  0.309185 -1.003  0.31574
## site32R  0.046774  1.047885  0.404984  0.115  0.90805
## site33R -0.257920  0.772657  0.354189 -0.728  0.46649
## site34R -0.233721  0.791583  0.275695 -0.848  0.39658
## site35R -0.171681  0.842248  0.275652 -0.623  0.53341
## site36R -0.357971  0.699093  0.309997 -1.155  0.24819
## site37R -0.222043  0.800881  0.310850 -0.714  0.47504
## site38R -0.398633  0.671237  0.301931 -1.320  0.18674
## site39R  0.089268  1.093374  0.312885  0.285  0.77541
## site40R  0.598770  1.819878  0.332727  1.800  0.07193
## site41R -0.155584  0.855915  0.286248 -0.544  0.58677
## site42R  0.033413  1.033977  0.325191  0.103  0.91816
## 
## Likelihood ratio test=180.4  on 48 df, p=< 2.2e-16
## n= 87154, number of events= 1785 
##    (472 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit1 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=PFF[,c(Xnm, 'start', 'end', 'event', 'dx_yr', 'site')], q=4)
qc.survfit1

## Scaled effect size (positive direction, sum of positive coefficients = 0.169)
##     BC    NH4     SS 
## 0.7680 0.1586 0.0735 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.29)
##     NO3     SO4    Soil      OM 
## 0.44279 0.37566 0.17900 0.00255 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error Lower CI Upper CI Z value Pr(>|z|)
## psi1 -0.120975   0.070184 -0.25853 0.016584 -1.7237  0.08477

#Lastly the HR is reported through the following
exp(qc.survfit1$coef)

##      psi1 
## 0.8860562

exp(qc.survfit1$ci)

## [1] 0.772183 1.016722

So the HR of the overall model is 0.90 (95% CI 0.78-1.03) and the positive direction of effect (i.e. harmful) primarily driven by BC, OM, and SS

Now to plot the findings

plot(qc.survfit1)

Complete multi-pollutant model + dx_yr

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + smokeHx + dich_Race + disadv + dx_yr + site, data=PFF)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + 
##     smokeHx + dich_Race + disadv + dx_yr + site, data = PFF)
## 
##                         coef exp(coef)  se(coef)      z        p
## SO4                 0.021920  1.022162  0.191678  0.114  0.90895
## NH4                 0.951012  2.588328  0.461999  2.058  0.03954
## NO3                -0.527852  0.589870  0.170432 -3.097  0.00195
## BC                 -0.438160  0.645222  0.370825 -1.182  0.23737
## OM                  0.195806  1.216291  0.078690  2.488  0.01283
## SS                  0.121016  1.128643  0.126009  0.960  0.33687
## Soil               -0.167273  0.845969  0.189865 -0.881  0.37831
## age_dx              0.001175  1.001176  0.002512  0.468  0.63988
## sexM                0.043783  1.044756  0.052731  0.830  0.40637
## smokeHxEver         0.045716  1.046777  0.050980  0.897  0.36986
## dich_RaceNon-White -0.043220  0.957701  0.082515 -0.524  0.60043
## disadv             -0.035087  0.965521  0.091537 -0.383  0.70149
## dx_yr               0.142622  1.153294  0.019498  7.315 2.58e-13
## site02R             0.629176  1.876065  0.302911  2.077  0.03779
## site03R             0.037433  1.038142  0.293576  0.128  0.89854
## site04R             0.079901  1.083180  0.337714  0.237  0.81297
## site05R            -0.443830  0.641574  0.347134 -1.279  0.20105
## site06R            -0.189483  0.827387  0.325640 -0.582  0.56065
## site07R            -0.234973  0.790592  0.303158 -0.775  0.43829
## site08R                   NA        NA  0.000000     NA       NA
## site09R            -0.128599  0.879327  0.294854 -0.436  0.66273
## site10R            -0.016857  0.983284  0.371045 -0.045  0.96376
## site11R            -0.277704  0.757521  0.310576 -0.894  0.37124
## site12R             0.117737  1.124949  0.296374  0.397  0.69118
## site13R            -0.362419  0.695990  0.292588 -1.239  0.21547
## site14R            -0.070755  0.931691  0.429353 -0.165  0.86911
## site15R            -0.140608  0.868830  0.323541 -0.435  0.66386
## site16R             0.069303  1.071761  0.294569  0.235  0.81400
## site17R            -0.018262  0.981904  0.335722 -0.054  0.95662
## site18R            -0.391164  0.676269  0.322482 -1.213  0.22514
## site19R             0.230148  1.258786  0.381643  0.603  0.54648
## site20R            -0.101366  0.903602  0.339560 -0.299  0.76530
## site21R            -0.444343  0.641245  0.320101 -1.388  0.16510
## site22R            -0.043102  0.957814  0.298223 -0.145  0.88508
## site23R            -0.154049  0.857230  0.333941 -0.461  0.64458
## site24R             0.177724  1.194496  0.296633  0.599  0.54908
## site25R             0.120417  1.127967  0.302658  0.398  0.69073
## site26R            -0.400965  0.669674  0.371986 -1.078  0.28108
## site27R            -0.150516  0.860264  0.406888 -0.370  0.71144
## site28R            -0.170347  0.843372  0.340805 -0.500  0.61719
## site29R            -0.346508  0.707153  0.363679 -0.953  0.34070
## site30R             0.076594  1.079604  0.332195  0.231  0.81765
## site31R            -0.323074  0.723920  0.314171 -1.028  0.30379
## site32R            -0.011359  0.988705  0.415500 -0.027  0.97819
## site33R            -0.302228  0.739170  0.360705 -0.838  0.40210
## site34R            -0.258191  0.772448  0.281290 -0.918  0.35868
## site35R            -0.219257  0.803115  0.282376 -0.776  0.43747
## site36R            -0.371007  0.690039  0.316293 -1.173  0.24080
## site37R            -0.232755  0.792348  0.316459 -0.735  0.46204
## site38R            -0.427690  0.652014  0.309347 -1.383  0.16680
## site39R             0.047296  1.048433  0.316894  0.149  0.88136
## site40R             0.560805  1.752083  0.338463  1.657  0.09754
## site41R            -0.158676  0.853273  0.291568 -0.544  0.58629
## site42R             0.021131  1.021355  0.329707  0.064  0.94890
## 
## Likelihood ratio test=181.3  on 53 df, p=6.214e-16
## n= 85935, number of events= 1759 
##    (1691 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit2 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=PFF[,c(Xnm, 'age_dx', 'sex', 'smokeHx', 'dich_Race', 'disadv', 'start', 'end', 'event', 'dx_yr', 'site')], q=4)
qc.survfit2

## Scaled effect size (positive direction, sum of positive coefficients = 0.168)
##     BC    NH4     SS     OM 
## 0.8067 0.0875 0.0547 0.0511 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.294)
##   NO3   SO4  Soil 
## 0.455 0.366 0.180 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error Lower CI Upper CI Z value Pr(>|z|)
## psi1 -0.125920   0.071084 -0.26524 0.013403 -1.7714  0.07649

#Lastly the HR is reported through the following
exp(qc.survfit2$coef)

##      psi1 
## 0.8816856

exp(qc.survfit2$ci)

## [1] 0.767020 1.013493

So the HR of the overall model is 0.90 (95% CI 0.79-1.03) and the positive direction of effect (i.e. harmful) primarily driven by BC, OM, and SS

Now to plot the findings

plot(qc.survfit2)

89.3 CARE Alone

Multi-pollutant model without other covariates aside from dx_yr and site

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site, data=CARE)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site, 
##     data = CARE)
## 
##              coef exp(coef)  se(coef)      z        p
## SO4      0.132154  1.141284  0.347663  0.380  0.70386
## NH4     -5.459114  0.004257  0.681654 -8.009 1.16e-15
## NO3      1.771168  5.877715  0.234817  7.543 4.60e-14
## BC       2.543283 12.721368  0.486065  5.232 1.67e-07
## OM      -0.716815  0.488305  0.074945 -9.565  < 2e-16
## SS      -0.642256  0.526104  0.250671 -2.562  0.01040
## Soil     1.905576  6.723277  0.279244  6.824 8.85e-12
## dx_yr    0.659633  1.934083  0.019572 33.703  < 2e-16
## site102 -0.135506  0.873274  0.152558 -0.888  0.37442
## site103  0.459785  1.583734  0.161234  2.852  0.00435
## site104  0.440262  1.553114  0.173002  2.545  0.01093
## site105  0.611000  1.842273  0.098370  6.211 5.26e-10
## site106  0.190342  1.209663  0.081239  2.343  0.01913
## site107  0.899358  2.458024  0.144935  6.205 5.46e-10
## site108 -0.879056  0.415175  0.159393 -5.515 3.49e-08
## 
## Likelihood ratio test=5081  on 15 df, p=< 2.2e-16
## n= 160073, number of events= 3301 
##    (142 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit1 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=CARE[,c(Xnm, 'start', 'end', 'event', 'dx_yr', 'site')], q=4)
qc.survfit1

## Scaled effect size (positive direction, sum of positive coefficients = 0.375)
##  Soil   NO3    BC 
## 0.386 0.318 0.296 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.647)
##    NH4     OM    SO4     SS 
## 0.4380 0.3815 0.1021 0.0784 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error Lower CI Upper CI Z value Pr(>|z|)
## psi1 -0.271654   0.046389 -0.36257 -0.18073  -5.856 4.74e-09

#Lastly the HR is reported through the following
exp(qc.survfit1$coef)

##      psi1 
## 0.7621178

exp(qc.survfit1$ci)

## [1] 0.6958825 0.8346574

So the HR of the overall model is 0.76 (95% CI 0.70-0.83) with the positive direction of effects (i.e. harmful) primarily driven by Soil, NO3, and BC

Now to plot the findings

plot(qc.survfit1)

Complete multi-pollutant model + dx_yr

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + smokeHx + dich_Race + disadv + dx_yr + site, data=CARE)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + 
##     smokeHx + dich_Race + disadv + dx_yr + site, data = CARE)
## 
##                         coef exp(coef)  se(coef)      z        p
## SO4                 0.243661  1.275912  0.352160  0.692  0.48900
## NH4                -5.667471  0.003457  0.685851 -8.263  < 2e-16
## NO3                 1.813419  6.131374  0.235893  7.687 1.50e-14
## BC                  2.632680 13.911004  0.493520  5.334 9.58e-08
## OM                 -0.723846  0.484884  0.075697 -9.562  < 2e-16
## SS                 -0.651174  0.521433  0.257022 -2.534  0.01129
## Soil                1.810289  6.112213  0.281468  6.432 1.26e-10
## age_dx              0.010119  1.010170  0.001615  6.264 3.74e-10
## sexF               -0.109652  0.896146  0.036632 -2.993  0.00276
## smokeHxFormer       0.025519  1.025848  0.038756  0.658  0.51024
## smokeHxAlways       0.003077  1.003082  0.083667  0.037  0.97066
## smokeHxUnknown      0.155587  1.168344  0.337416  0.461  0.64472
## dich_RaceNon-White -0.027267  0.973101  0.047657 -0.572  0.56721
## disadv             -0.067314  0.934902  0.066452 -1.013  0.31107
## dx_yr               0.655309  1.925737  0.019752 33.176  < 2e-16
## site102            -0.124368  0.883055  0.154813 -0.803  0.42178
## site103             0.479227  1.614826  0.163256  2.935  0.00333
## site104             0.433483  1.542621  0.174896  2.479  0.01319
## site105             0.592685  1.808839  0.099434  5.961 2.51e-09
## site106             0.109507  1.115728  0.082152  1.333  0.18254
## site107             0.866340  2.378190  0.146889  5.898 3.68e-09
## site108            -0.909239  0.402831  0.160938 -5.650 1.61e-08
## 
## Likelihood ratio test=5144  on 22 df, p=< 2.2e-16
## n= 160069, number of events= 3300 
##    (146 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit2 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=CARE[,c(Xnm, 'age_dx', 'sex', 'smokeHx', 'dich_Race', 'disadv', 'start', 'end', 'event', 'dx_yr', 'site')], q=4)
qc.survfit2

## Scaled effect size (positive direction, sum of positive coefficients = 0.378)
##  Soil   NO3    BC 
## 0.390 0.317 0.293 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.647)
##    NH4     OM    SO4     SS 
## 0.4449 0.3746 0.0912 0.0893 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error Lower CI Upper CI Z value  Pr(>|z|)
## psi1 -0.268395   0.046872 -0.36026 -0.17653 -5.7261 1.027e-08

#Lastly the HR is reported through the following
exp(qc.survfit2$coef)

##      psi1 
## 0.7646059

exp(qc.survfit2$ci)

## [1] 0.6974935 0.8381757

So the HR of the overall model is 0.76 (95% CI 0.70-0.84) with the positive direction of effects (i.e. harmful) primarily driven by Soil, NO3, and BC

Now to plot the findings

plot(qc.survfit2)

89.4 One-Stage Meta-Analysis - i.e. Combined Cohorts

Multi-pollutant model without other covariates aside from dx_yr and site

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site + cluster(cohort), data=All)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site, 
##     data = All, cluster = cohort)
## 
##              coef exp(coef)  se(coef) robust se      z        p
## SO4      1.108729  3.030505  0.086215  0.538426  2.059 0.039475
## NH4     -0.993313  0.370348  0.265729  1.548234 -0.642 0.521147
## NO3     -0.154710  0.856663  0.113661  0.476470 -0.325 0.745408
## BC       3.006893 20.224456  0.189290  2.362709  1.273 0.203144
## OM      -0.529659  0.588806  0.038869  0.409164 -1.294 0.195496
## SS      -0.063886  0.938112  0.103118  0.379890 -0.168 0.866450
## Soil    -0.121733  0.885384  0.132499  0.287495 -0.423 0.671983
## dx_yr    0.350893  1.420336  0.009046  0.148491  2.363 0.018125
## site02R  0.410522  1.507605  0.286724  0.389285  1.055 0.291629
## site03R -0.535539  0.585353  0.275947  0.515505 -1.039 0.298868
## site04R -1.463077  0.231523  0.296217  0.883062 -1.657 0.097555
## site05R -1.894651  0.150371  0.288781  0.786802 -2.408 0.016038
## site06R -1.437573  0.237504  0.276405  0.637345 -2.256 0.024098
## site07R -1.362328  0.256064  0.262883  0.647437 -2.104 0.035362
## site09R -0.966249  0.380508  0.273584  0.414404 -2.332 0.019719
## site1   -2.284759  0.101799  0.230902  1.091159 -2.094 0.036270
## site101 -1.113484  0.328413  0.227134  0.503861 -2.210 0.027112
## site102 -0.374908  0.687353  0.222824  0.176022 -2.130 0.033180
## site103 -0.208375  0.811902  0.215863  0.088923 -2.343 0.019113
## site104 -0.393795  0.674492  0.222376  0.054787 -7.188 6.59e-13
## site105 -0.628453  0.533416  0.220729  0.176534 -3.560 0.000371
## site106 -1.237766  0.290031  0.228386  0.551667 -2.244 0.024853
## site107 -0.347693  0.706316  0.243717  0.181918 -1.911 0.055971
## site108 -0.447822  0.639018  0.245688  0.430183 -1.041 0.297874
## site10R -1.205880  0.299428  0.317410  0.494673 -2.438 0.014780
## site11R -1.683655  0.185694  0.253589  0.780840 -2.156 0.031067
## site12R -0.617754  0.539154  0.267632  0.437462 -1.412 0.157912
## site13R -1.902421  0.149207  0.251546  0.807326 -2.356 0.018451
## site14R -1.313274  0.268938  0.405549  0.643241 -2.042 0.041186
## site15R -1.463415  0.231445  0.279684  0.597869 -2.448 0.014376
## site16R -1.024201  0.359083  0.272177  0.549982 -1.862 0.062569
## site17R -1.315865  0.268242  0.291586  0.717923 -1.833 0.066821
## site18R -1.734137  0.176552  0.272034  0.674880 -2.570 0.010183
## site19R -0.422756  0.655239  0.330864  0.634581 -0.666 0.505286
## site20R -1.136242  0.321023  0.302036  0.602719 -1.885 0.059404
## site21R -2.157402  0.115625  0.268768  0.930645 -2.318 0.020440
## site22R -1.541335  0.214095  0.262533  0.826630 -1.865 0.062237
## site23R -1.265254  0.282168  0.274567  0.665038 -1.903 0.057102
## site24R -0.425133  0.653683  0.266214  0.378451 -1.123 0.261288
## site25R -0.129724  0.878337  0.280325  0.315273 -0.411 0.680730
## site26R -1.732274  0.176882  0.298739  0.695066 -2.492 0.012694
## site27R -1.663569  0.189462  0.376894  0.720590 -2.309 0.020965
## site28R -1.822800  0.161573  0.282450  0.877685 -2.077 0.037818
## site29R -1.537592  0.214898  0.341795  0.517310 -2.972 0.002956
## site30R -1.653168  0.191442  0.286892  1.041288 -1.588 0.112373
## site31R -1.618317  0.198232  0.277169  0.618423 -2.617 0.008875
## site32R -0.854110  0.425662  0.341808  0.881196 -0.969 0.332415
## site33R -1.565045  0.209079  0.289238  0.701808 -2.230 0.025746
## site34R -1.196584  0.302225  0.253687  0.511347 -2.340 0.019280
## site35R -0.784158  0.456504  0.266177  0.249101 -3.148 0.001644
## site36R -1.694074  0.183769  0.268098  0.799006 -2.120 0.033987
## site37R -1.721940  0.178719  0.273122  0.823633 -2.091 0.036558
## site38R -1.667340  0.188749  0.267896  0.584201 -2.854 0.004317
## site39R -0.445251  0.640664  0.304685  0.529867 -0.840 0.400737
## site40R -1.296523  0.273481  0.317782  1.576595 -0.822 0.410874
## site41R -1.543715  0.213586  0.264996  0.786001 -1.964 0.049529
## site42R -1.468058  0.230372  0.287490  0.735193 -1.997 0.045843
## 
## Likelihood ratio test=3375  on 57 df, p=< 2.2e-16
## n= 335367, number of events= 6459 
##    (1981 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit1 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=All[,c(Xnm, 'start', 'end', 'event', 'dx_yr', 'site', 'cohort')], q=4)
qc.survfit1

## Scaled effect size (positive direction, sum of positive coefficients = 0.592)
##    BC  Soil   SO4 
## 0.600 0.229 0.172 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.587)
##     OM    NH4    NO3     SS 
## 0.4111 0.3254 0.2333 0.0301 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error  Lower CI Upper CI Z value Pr(>|z|)
## psi1 0.0048171  0.0374287 -0.068542 0.078176  0.1287   0.8976

#Lastly the HR is reported through the following
exp(qc.survfit1$coef)

##     psi1 
## 1.004829

exp(qc.survfit1$ci)

## [1] 0.9337545 1.0813129

So the HR of the overall model is 1.02 (95% CI 0.95-1.09) with the positive direction of effects (i.e. harmful) primarily driven by BC, Soil, SO4

Now to plot the findings

plot(qc.survfit1)

Complete multi-pollutant model + dx_yr

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + smokeHx + dich_Race + disadv + dx_yr + site + cluster(cohort), data=All)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + 
##     smokeHx + dich_Race + disadv + dx_yr + site, data = All, 
##     cluster = cohort)
## 
##                         coef exp(coef)  se(coef) robust se       z        p
## SO4                 1.108152  3.028756  0.087559  0.558069   1.986 0.047068
## NH4                -0.977642  0.376197  0.269984  1.614064  -0.606 0.544713
## NO3                -0.145273  0.864787  0.115270  0.509733  -0.285 0.775646
## BC                  3.147118 23.268916  0.191483  2.373037   1.326 0.184774
## OM                 -0.550872  0.576447  0.039446  0.410813  -1.341 0.179944
## SS                 -0.093092  0.911110  0.104288  0.393518  -0.237 0.812996
## Soil               -0.125732  0.881852  0.133200  0.263884  -0.476 0.633743
## age_dx              0.008946  1.008987  0.001176  0.003174   2.819 0.004822
## sexF               -0.150299  0.860451  0.026733  0.065005  -2.312 0.020772
## smokeHxFormer       0.105799  1.111599  0.033080  0.038742   2.731 0.006317
## smokeHxAlways      -0.011978  0.988094  0.075048  0.043948  -0.273 0.785205
## smokeHxUnknown     -0.109979  0.895853  0.070515  0.047446  -2.318 0.020452
## smokeHxEver         0.001228  1.001229  0.049880  0.017999   0.068 0.945603
## dich_RaceNon-White  0.008509  1.008545  0.037070  0.039067   0.218 0.827586
## disadv              0.100696  1.105940  0.046483  0.046936   2.145 0.031923
## dx_yr               0.348953  1.417583  0.009249  0.150440   2.320 0.020365
## site02R             0.414442  1.513526  0.291126  0.375301   1.104 0.269467
## site03R            -0.523422  0.592489  0.280146  0.501853  -1.043 0.296958
## site04R            -1.459834  0.232275  0.300887  0.911960  -1.601 0.109429
## site05R            -1.925254  0.145839  0.294909  0.788842  -2.441 0.014663
## site06R            -1.530439  0.216441  0.282460  0.629291  -2.432 0.015015
## site07R            -1.436115  0.237850  0.267144  0.655942  -2.189 0.028568
## site09R            -1.041338  0.352982  0.279786  0.421492  -2.471 0.013489
## site1              -2.372598  0.093238  0.238308  1.071202  -2.215 0.026767
## site101            -1.150650  0.316431  0.233772  0.482478  -2.385 0.017085
## site102            -0.397530  0.671978  0.229572  0.152994  -2.598 0.009368
## site103            -0.248863  0.779687  0.223122  0.056570  -4.399 1.09e-05
## site104            -0.452037  0.636331  0.229292  0.034147 -13.238  < 2e-16
## site105            -0.675072  0.509120  0.228031  0.151917  -4.444 8.84e-06
## site106            -1.355233  0.257887  0.235201  0.528953  -2.562 0.010404
## site107            -0.398735  0.671169  0.250199  0.164161  -2.429 0.015143
## site108            -0.511230  0.599757  0.251762  0.412234  -1.240 0.214922
## site10R            -1.246929  0.287386  0.324246  0.450080  -2.770 0.005598
## site11R            -1.785533  0.167708  0.260170  0.768710  -2.323 0.020192
## site12R            -0.695944  0.498603  0.271990  0.466269  -1.493 0.135547
## site13R            -2.007853  0.134277  0.256179  0.817602  -2.456 0.014058
## site14R            -1.331048  0.264200  0.408305  0.642390  -2.072 0.038263
## site15R            -1.552995  0.211613  0.284675  0.587252  -2.645 0.008181
## site16R            -1.039581  0.353603  0.276552  0.552473  -1.882 0.059878
## site17R            -1.331872  0.263983  0.296084  0.725862  -1.835 0.066523
## site18R            -1.798532  0.165542  0.276636  0.644237  -2.792 0.005243
## site19R            -0.479149  0.619310  0.335010  0.588349  -0.814 0.415418
## site20R            -1.183920  0.306076  0.305999  0.611662  -1.936 0.052919
## site21R            -2.236702  0.106810  0.273614  0.919152  -2.433 0.014956
## site22R            -1.636044  0.194749  0.267436  0.842882  -1.941 0.052257
## site23R            -1.315081  0.268453  0.280580  0.630726  -2.085 0.037067
## site24R            -0.426788  0.652602  0.270521  0.394885  -1.081 0.279791
## site25R            -0.143846  0.866021  0.286185  0.294590  -0.488 0.625343
## site26R            -1.794732  0.166172  0.302462  0.644882  -2.783 0.005385
## site27R            -1.679963  0.186381  0.379947  0.732014  -2.295 0.021734
## site28R            -1.823120  0.161521  0.287391  0.883932  -2.063 0.039159
## site29R            -1.561639  0.209792  0.345921  0.524446  -2.978 0.002904
## site30R            -1.771261  0.170118  0.292786  1.076019  -1.646 0.099738
## site31R            -1.642615  0.193474  0.281385  0.619613  -2.651 0.008025
## site32R            -0.880623  0.414525  0.348876  0.810909  -1.086 0.277492
## site33R            -1.609281  0.200031  0.293709  0.640906  -2.511 0.012041
## site34R            -1.257423  0.284386  0.258531  0.520677  -2.415 0.015736
## site35R            -0.845418  0.429378  0.271976  0.248153  -3.407 0.000657
## site36R            -1.759497  0.172131  0.272735  0.808631  -2.176 0.029563
## site37R            -1.814536  0.162913  0.277593  0.846376  -2.144 0.032042
## site38R            -1.724598  0.178245  0.273644  0.582442  -2.961 0.003067
## site39R            -0.528256  0.589632  0.309102  0.530223  -0.996 0.319109
## site40R            -1.458496  0.232586  0.324919  1.611728  -0.905 0.365504
## site41R            -1.608660  0.200156  0.269845  0.803091  -2.003 0.045168
## site42R            -1.510156  0.220875  0.291605  0.739732  -2.041 0.041202
## 
## Likelihood ratio test=3543  on 65 df, p=< 2.2e-16
## n= 330899, number of events= 6391 
##    (6449 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit2 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=All[,c(Xnm, 'age_dx', 'sex', 'smokeHx', 'dich_Race', 'disadv', 'start', 'end', 'event', 'dx_yr', 'site', 'cohort')], q=4)
qc.survfit2

## Scaled effect size (positive direction, sum of positive coefficients = 0.6)
##    BC  Soil   SO4 
## 0.590 0.212 0.198 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.589)
##     OM    NH4    NO3     SS 
## 0.3921 0.3252 0.2386 0.0441 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##      Estimate Std. Error  Lower CI Upper CI Z value Pr(>|z|)
## psi1 0.011551   0.037929 -0.062789 0.085891  0.3045   0.7607

#Lastly the HR is reported through the following
exp(qc.survfit2$coef)

##     psi1 
## 1.011618

exp(qc.survfit2$ci)

## [1] 0.9391415 1.0896873

So the HR of the overall model is 1.02 (95% CI 0.95-1.10) with the positive direction of effects (i.e. harmful) primarily driven by BC, SO4, and Soil

Now to plot the findings

plot(qc.survfit2)

90 Creating IPF-specific Dataframes

90.1 IPF-Only Simmons

Simm_IPF <- Simm %>% filter(dx=="IPF")

90.2 IPF-Only PFF

PFF_IPF <- PFF %>% filter(dx=="IPF")

90.3 IPF-Only CARE-PF

CARE_IPF <- CARE %>% filter(dx=="IPF")

90.4 IPF-Only Combined Cohorts

All_IPF <- All %>% filter(dx=="IPF")

91 IPF-Alone Subgroup Analysis with Time-Weighted Exposure Estimates

91.1 IPF-Only PM2.5

91.1.1 IPF-Only PM2.5 - Simmons Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ PM + dx_yr, data=Simm_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM + dx_yr, data = Simm_IPF, 
##     id = ID)
## 
##   n= 31982, number of events= 695 
##    (446 observations deleted due to missingness)
## 
##          coef exp(coef) se(coef)     z Pr(>|z|)   
## PM    0.02683   1.02719  0.02436 1.101  0.27079   
## dx_yr 0.03347   1.03404  0.01249 2.681  0.00734 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##       exp(coef) exp(-coef) lower .95 upper .95
## PM        1.027     0.9735    0.9793     1.077
## dx_yr     1.034     0.9671    1.0090     1.060
## 
## Concordance= 0.52  (se = 0.013 )
## Likelihood ratio test= 8.81  on 2 df,   p=0.01
## Wald test            = 8.75  on 2 df,   p=0.01
## Score (logrank) test = 8.76  on 2 df,   p=0.01

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ PM + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv, data=Simm_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv, data = Simm_IPF, id = ID)
## 
##   n= 30488, number of events= 670 
##    (1940 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## PM                  0.026844  1.027208  0.025293  1.061  0.28854    
## dx_yr               0.021340  1.021570  0.013194  1.617  0.10579    
## age_dx              0.002010  1.002012  0.004555  0.441  0.65900    
## sexF               -0.375968  0.686625  0.088853 -4.231 2.32e-05 ***
## dich_RaceNon-White  0.163702  1.177863  0.128527  1.274  0.20278    
## smokeHxFormer       0.082335  1.085820  0.091977  0.895  0.37069    
## smokeHxAlways      -0.402321  0.668766  0.266624 -1.509  0.13131    
## smokeHxUnknown      0.495516  1.641345  0.169973  2.915  0.00355 ** 
## disadv              0.385780  1.470761  0.134234  2.874  0.00405 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## PM                    1.0272     0.9735    0.9775    1.0794
## dx_yr                 1.0216     0.9789    0.9955    1.0483
## age_dx                1.0020     0.9980    0.9931    1.0110
## sexF                  0.6866     1.4564    0.5769    0.8172
## dich_RaceNon-White    1.1779     0.8490    0.9156    1.5153
## smokeHxFormer         1.0858     0.9210    0.9067    1.3003
## smokeHxAlways         0.6688     1.4953    0.3966    1.1278
## smokeHxUnknown        1.6413     0.6093    1.1763    2.2902
## disadv                1.4708     0.6799    1.1305    1.9134
## 
## Concordance= 0.591  (se = 0.012 )
## Likelihood ratio test= 52.75  on 9 df,   p=3e-08
## Wald test            = 53.33  on 9 df,   p=3e-08
## Score (logrank) test = 54.06  on 9 df,   p=2e-08

91.1.2 IPF-Only PM2.5 - PFF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ PM + dx_yr + site, data=PFF_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM + dx_yr + site, 
##     data = PFF_IPF, id = ID)
## 
##   n= 54499, number of events= 1146 
##    (217 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef)      z Pr(>|z|)    
## PM       0.011916  1.011987  0.025653  0.464   0.6423    
## dx_yr    0.107435  1.113418  0.015228  7.055 1.73e-12 ***
## site02R  0.623805  1.866015  0.344110  1.813   0.0699 .  
## site03R -0.242860  0.784381  0.347132 -0.700   0.4842    
## site04R -0.249013  0.779570  0.458674 -0.543   0.5872    
## site05R -0.349963  0.704714  0.385035 -0.909   0.3634    
## site06R -0.052940  0.948437  0.331987 -0.159   0.8733    
## site07R -0.333392  0.716489  0.312225 -1.068   0.2856    
## site08R        NA        NA  0.000000     NA       NA    
## site09R -0.032164  0.968347  0.342388 -0.094   0.9252    
## site10R -0.158498  0.853425  0.386348 -0.410   0.6816    
## site11R -0.093865  0.910405  0.310083 -0.303   0.7621    
## site12R -0.088763  0.915062  0.334528 -0.265   0.7907    
## site13R -0.314391  0.730234  0.308481 -1.019   0.3081    
## site14R -0.382317  0.682278  0.645221 -0.593   0.5535    
## site15R  0.015737  1.015861  0.335894  0.047   0.9626    
## site16R  0.117281  1.124435  0.348898  0.336   0.7368    
## site17R  0.001297  1.001297  0.371099  0.003   0.9972    
## site18R -0.296922  0.743102  0.329973 -0.900   0.3682    
## site19R -0.001157  0.998843  0.350358 -0.003   0.9974    
## site20R -0.265193  0.767058  0.350537 -0.757   0.4493    
## site21R -0.321279  0.725221  0.332269 -0.967   0.3336    
## site22R -0.089091  0.914762  0.314649 -0.283   0.7771    
## site23R -0.179658  0.835556  0.328299 -0.547   0.5842    
## site24R -0.147814  0.862592  0.332239 -0.445   0.6564    
## site25R -0.032406  0.968114  0.330662 -0.098   0.9219    
## site26R -0.418976  0.657720  0.348948 -1.201   0.2299    
## site27R -0.008847  0.991192  0.762873 -0.012   0.9907    
## site28R -0.160623  0.851613  0.447259 -0.359   0.7195    
## site29R -0.294579  0.744845  0.436056 -0.676   0.4993    
## site30R -0.250205  0.778641  0.349491 -0.716   0.4740    
## site31R -0.319448  0.726550  0.450007 -0.710   0.4778    
## site32R -0.363701  0.695099  0.371617 -0.979   0.3277    
## site33R -0.352929  0.702627  0.336332 -1.049   0.2940    
## site34R -0.148796  0.861745  0.314529 -0.473   0.6362    
## site35R -0.109250  0.896506  0.334037 -0.327   0.7436    
## site36R -0.228369  0.795830  0.348053 -0.656   0.5117    
## site37R -0.197537  0.820750  0.344484 -0.573   0.5664    
## site38R -0.301230  0.739907  0.333182 -0.904   0.3659    
## site39R -0.208823  0.811539  0.360762 -0.579   0.5627    
## site40R  0.094558  1.099173  0.383902  0.246   0.8054    
## site41R -0.226236  0.797530  0.373636 -0.605   0.5448    
## site42R  0.048681  1.049886  0.368957  0.132   0.8950    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## PM         1.0120     0.9882    0.9624     1.064
## dx_yr      1.1134     0.8981    1.0807     1.147
## site02R    1.8660     0.5359    0.9506     3.663
## site03R    0.7844     1.2749    0.3972     1.549
## site04R    0.7796     1.2828    0.3173     1.915
## site05R    0.7047     1.4190    0.3313     1.499
## site06R    0.9484     1.0544    0.4948     1.818
## site07R    0.7165     1.3957    0.3885     1.321
## site08R        NA         NA        NA        NA
## site09R    0.9683     1.0327    0.4950     1.894
## site10R    0.8534     1.1717    0.4002     1.820
## site11R    0.9104     1.0984    0.4958     1.672
## site12R    0.9151     1.0928    0.4750     1.763
## site13R    0.7302     1.3694    0.3989     1.337
## site14R    0.6823     1.4657    0.1926     2.416
## site15R    1.0159     0.9844    0.5259     1.962
## site16R    1.1244     0.8893    0.5675     2.228
## site17R    1.0013     0.9987    0.4838     2.072
## site18R    0.7431     1.3457    0.3892     1.419
## site19R    0.9988     1.0012    0.5027     1.985
## site20R    0.7671     1.3037    0.3859     1.525
## site21R    0.7252     1.3789    0.3781     1.391
## site22R    0.9148     1.0932    0.4937     1.695
## site23R    0.8356     1.1968    0.4391     1.590
## site24R    0.8626     1.1593    0.4498     1.654
## site25R    0.9681     1.0329    0.5064     1.851
## site26R    0.6577     1.5204    0.3319     1.303
## site27R    0.9912     1.0089    0.2222     4.421
## site28R    0.8516     1.1742    0.3544     2.046
## site29R    0.7448     1.3426    0.3169     1.751
## site30R    0.7786     1.2843    0.3925     1.545
## site31R    0.7266     1.3764    0.3008     1.755
## site32R    0.6951     1.4386    0.3355     1.440
## site33R    0.7026     1.4232    0.3634     1.358
## site34R    0.8617     1.1604    0.4652     1.596
## site35R    0.8965     1.1154    0.4658     1.725
## site36R    0.7958     1.2565    0.4023     1.574
## site37R    0.8207     1.2184    0.4178     1.612
## site38R    0.7399     1.3515    0.3851     1.422
## site39R    0.8115     1.2322    0.4002     1.646
## site40R    1.0992     0.9098    0.5180     2.333
## site41R    0.7975     1.2539    0.3834     1.659
## site42R    1.0499     0.9525    0.5094     2.164
## 
## Concordance= 0.59  (se = 0.009 )
## Likelihood ratio test= 110.7  on 42 df,   p=4e-08
## Wald test            = 109.1  on 42 df,   p=7e-08
## Score (logrank) test = 113.4  on 42 df,   p=2e-08

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ PM + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=PFF_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = PFF_IPF, 
##     id = ID)
## 
##   n= 53867, number of events= 1131 
##    (849 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)      z Pr(>|z|)    
## PM                  0.0142839  1.0143864  0.0261075  0.547    0.584    
## dx_yr               0.1068047  1.1127170  0.0157189  6.795 1.09e-11 ***
## age_dx             -0.0002805  0.9997195  0.0039344 -0.071    0.943    
## sexM                0.0611311  1.0630382  0.0713103  0.857    0.391    
## dich_RaceNon-White -0.0625902  0.9393284  0.1263612 -0.495    0.620    
## smokeHxEver         0.0430378  1.0439774  0.0646553  0.666    0.506    
## disadv             -0.0517598  0.9495569  0.1120933 -0.462    0.644    
## site02R             0.5681626  1.7650210  0.3575818  1.589    0.112    
## site03R            -0.2710643  0.7625675  0.3591658 -0.755    0.450    
## site04R            -0.3103185  0.7332134  0.4702289 -0.660    0.509    
## site05R            -0.4026734  0.6685304  0.3987572 -1.010    0.313    
## site06R            -0.0940045  0.9102787  0.3484233 -0.270    0.787    
## site07R            -0.3807946  0.6833182  0.3272209 -1.164    0.245    
## site08R                    NA         NA  0.0000000     NA       NA    
## site09R            -0.1151220  0.8912574  0.3573719 -0.322    0.747    
## site10R            -0.2176135  0.8044363  0.4001451 -0.544    0.587    
## site11R            -0.1573101  0.8544391  0.3293543 -0.478    0.633    
## site12R            -0.1341212  0.8744841  0.3486543 -0.385    0.700    
## site13R            -0.3756688  0.6868297  0.3226027 -1.164    0.244    
## site14R            -0.4489588  0.6382924  0.6532020 -0.687    0.492    
## site15R            -0.0369873  0.9636883  0.3523427 -0.105    0.916    
## site16R             0.0691054  1.0715492  0.3608654  0.191    0.848    
## site17R            -0.0625400  0.9393755  0.3857200 -0.162    0.871    
## site18R            -0.3363485  0.7143741  0.3464050 -0.971    0.332    
## site19R            -0.0457425  0.9552879  0.3625823 -0.126    0.900    
## site20R            -0.3202992  0.7259318  0.3650363 -0.877    0.380    
## site21R            -0.3713135  0.6898276  0.3472106 -1.069    0.285    
## site22R            -0.1477421  0.8626536  0.3299279 -0.448    0.654    
## site23R            -0.2480252  0.7803403  0.3449457 -0.719    0.472    
## site24R            -0.2052051  0.8144802  0.3463544 -0.592    0.554    
## site25R            -0.0686836  0.9336221  0.3444899 -0.199    0.842    
## site26R            -0.4564070  0.6335559  0.3626695 -1.258    0.208    
## site27R            -0.1192351  0.8875991  0.7742839 -0.154    0.878    
## site28R            -0.2003847  0.8184158  0.4602171 -0.435    0.663    
## site29R            -0.3512994  0.7037730  0.4479388 -0.784    0.433    
## site30R            -0.3135246  0.7308664  0.3639645 -0.861    0.389    
## site31R            -0.3707204  0.6902369  0.4592058 -0.807    0.419    
## site32R            -0.4300017  0.6505080  0.3952399 -1.088    0.277    
## site33R            -0.3966719  0.6725547  0.3523673 -1.126    0.260    
## site34R            -0.1931466  0.8243611  0.3301659 -0.585    0.559    
## site35R            -0.1699420  0.8437138  0.3485186 -0.488    0.626    
## site36R            -0.2670486  0.7656359  0.3636977 -0.734    0.463    
## site37R            -0.2471685  0.7810091  0.3584277 -0.690    0.490    
## site38R            -0.3422450  0.7101742  0.3474319 -0.985    0.325    
## site39R            -0.2614249  0.7699537  0.3735101 -0.700    0.484    
## site40R             0.0501533  1.0514323  0.3963693  0.127    0.899    
## site41R            -0.2687248  0.7643536  0.3846476 -0.699    0.485    
## site42R            -0.0227469  0.9775099  0.3821024 -0.060    0.953    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## PM                    1.0144     0.9858    0.9638     1.068
## dx_yr                 1.1127     0.8987    1.0790     1.148
## age_dx                0.9997     1.0003    0.9920     1.007
## sexM                  1.0630     0.9407    0.9244     1.222
## dich_RaceNon-White    0.9393     1.0646    0.7333     1.203
## smokeHxEver           1.0440     0.9579    0.9197     1.185
## disadv                0.9496     1.0531    0.7623     1.183
## site02R               1.7650     0.5666    0.8757     3.557
## site03R               0.7626     1.3114    0.3772     1.542
## site04R               0.7332     1.3639    0.2917     1.843
## site05R               0.6685     1.4958    0.3060     1.461
## site06R               0.9103     1.0986    0.4598     1.802
## site07R               0.6833     1.4634    0.3598     1.298
## site08R                   NA         NA        NA        NA
## site09R               0.8913     1.1220    0.4424     1.796
## site10R               0.8044     1.2431    0.3672     1.762
## site11R               0.8544     1.1704    0.4481     1.629
## site12R               0.8745     1.1435    0.4415     1.732
## site13R               0.6868     1.4560    0.3650     1.293
## site14R               0.6383     1.5667    0.1774     2.296
## site15R               0.9637     1.0377    0.4831     1.922
## site16R               1.0715     0.9332    0.5283     2.174
## site17R               0.9394     1.0645    0.4411     2.001
## site18R               0.7144     1.3998    0.3623     1.409
## site19R               0.9553     1.0468    0.4694     1.944
## site20R               0.7259     1.3775    0.3550     1.485
## site21R               0.6898     1.4496    0.3493     1.362
## site22R               0.8627     1.1592    0.4519     1.647
## site23R               0.7803     1.2815    0.3969     1.534
## site24R               0.8145     1.2278    0.4131     1.606
## site25R               0.9336     1.0711    0.4753     1.834
## site26R               0.6336     1.5784    0.3112     1.290
## site27R               0.8876     1.1266    0.1946     4.048
## site28R               0.8184     1.2219    0.3321     2.017
## site29R               0.7038     1.4209    0.2925     1.693
## site30R               0.7309     1.3682    0.3581     1.492
## site31R               0.6902     1.4488    0.2806     1.698
## site32R               0.6505     1.5373    0.2998     1.412
## site33R               0.6726     1.4869    0.3371     1.342
## site34R               0.8244     1.2131    0.4316     1.575
## site35R               0.8437     1.1852    0.4261     1.671
## site36R               0.7656     1.3061    0.3754     1.562
## site37R               0.7810     1.2804    0.3869     1.577
## site38R               0.7102     1.4081    0.3594     1.403
## site39R               0.7700     1.2988    0.3703     1.601
## site40R               1.0514     0.9511    0.4835     2.287
## site41R               0.7644     1.3083    0.3597     1.624
## site42R               0.9775     1.0230    0.4622     2.067
## 
## Concordance= 0.591  (se = 0.009 )
## Likelihood ratio test= 111.5  on 47 df,   p=4e-07
## Wald test            = 110.2  on 47 df,   p=5e-07
## Score (logrank) test = 114.5  on 47 df,   p=1e-07

91.1.3 IPF-Only PM2.5 - CARE-PF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ PM + dx_yr + site, data=CARE_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM + dx_yr + site, 
##     data = CARE_IPF, id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef)      z Pr(>|z|)    
## PM      -0.02785   0.97253  0.02487 -1.120 0.262773    
## dx_yr    0.81343   2.25563  0.03231 25.175  < 2e-16 ***
## site102  0.17913   1.19618  0.15161  1.182 0.237393    
## site103  0.47621   1.60996  0.12588  3.783 0.000155 ***
## site104  0.33123   1.39268  0.14417  2.297 0.021591 *  
## site105  0.05790   1.05961  0.12611  0.459 0.646148    
## site106  0.15043   1.16234  0.12170  1.236 0.216443    
## site107  0.10144   1.10676  0.19748  0.514 0.607493    
## site108 -0.43257   0.64884  0.26539 -1.630 0.103118    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## PM         0.9725     1.0282    0.9263     1.021
## dx_yr      2.2556     0.4433    2.1172     2.403
## site102    1.1962     0.8360    0.8887     1.610
## site103    1.6100     0.6211    1.2580     2.060
## site104    1.3927     0.7180    1.0499     1.847
## site105    1.0596     0.9437    0.8276     1.357
## site106    1.1623     0.8603    0.9157     1.475
## site107    1.1068     0.9035    0.7515     1.630
## site108    0.6488     1.5412    0.3857     1.092
## 
## Concordance= 0.774  (se = 0.01 )
## Likelihood ratio test= 1053  on 9 df,   p=<2e-16
## Wald test            = 667.6  on 9 df,   p=<2e-16
## Score (logrank) test = 586.3  on 9 df,   p=<2e-16

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ PM + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=CARE_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = CARE_IPF, 
##     id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## PM                 -0.033485  0.967069  0.025449 -1.316 0.188246    
## dx_yr               0.812722  2.254036  0.032467 25.032  < 2e-16 ***
## age_dx              0.012291  1.012367  0.004506  2.727 0.006382 ** 
## sexF               -0.043222  0.957699  0.077049 -0.561 0.574823    
## dich_RaceNon-White -0.145971  0.864183  0.110402 -1.322 0.186111    
## smokeHxFormer       0.001762  1.001764  0.081519  0.022 0.982751    
## smokeHxAlways       0.059700  1.061518  0.168799  0.354 0.723583    
## smokeHxUnknown      0.421664  1.524495  0.734804  0.574 0.566073    
## disadv             -0.009229  0.990814  0.129610 -0.071 0.943235    
## site102             0.151231  1.163265  0.154045  0.982 0.326233    
## site103             0.483593  1.621891  0.129013  3.748 0.000178 ***
## site104             0.331633  1.393242  0.145263  2.283 0.022431 *  
## site105             0.074673  1.077531  0.129274  0.578 0.563512    
## site106             0.122846  1.130710  0.122826  1.000 0.317231    
## site107             0.037946  1.038675  0.204876  0.185 0.853060    
## site108            -0.435598  0.646878  0.268748 -1.621 0.105052    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## PM                    0.9671     1.0341    0.9200     1.017
## dx_yr                 2.2540     0.4436    2.1151     2.402
## age_dx                1.0124     0.9878    1.0035     1.021
## sexF                  0.9577     1.0442    0.8235     1.114
## dich_RaceNon-White    0.8642     1.1572    0.6960     1.073
## smokeHxFormer         1.0018     0.9982    0.8538     1.175
## smokeHxAlways         1.0615     0.9420    0.7625     1.478
## smokeHxUnknown        1.5245     0.6560    0.3611     6.436
## disadv                0.9908     1.0093    0.7685     1.277
## site102               1.1633     0.8596    0.8601     1.573
## site103               1.6219     0.6166    1.2595     2.089
## site104               1.3932     0.7178    1.0480     1.852
## site105               1.0775     0.9280    0.8364     1.388
## site106               1.1307     0.8844    0.8888     1.438
## site107               1.0387     0.9628    0.6952     1.552
## site108               0.6469     1.5459    0.3820     1.095
## 
## Concordance= 0.776  (se = 0.01 )
## Likelihood ratio test= 1065  on 16 df,   p=<2e-16
## Wald test            = 674.9  on 16 df,   p=<2e-16
## Score (logrank) test = 593.2  on 16 df,   p=<2e-16

91.1.4 IPF-Only PM2.5 - Combined Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ PM + dx_yr + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM + dx_yr + site, 
##     data = All_IPF, id = ID, cluster = cohort)
## 
##   n= 123367, number of events= 2749 
##    (723 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef) robust se       z Pr(>|z|)    
## PM       0.08938   1.09349  0.01272   0.02513   3.557 0.000376 ***
## dx_yr    0.13165   1.14071  0.00799   0.07410   1.777 0.075627 .  
## site02R  0.39000   1.47698  0.33677   0.11493   3.393 0.000690 ***
## site03R -0.26172   0.76972  0.34707   0.02219 -11.796  < 2e-16 ***
## site04R -0.54224   0.58144  0.45158   0.13461  -4.028 5.62e-05 ***
## site05R -0.71039   0.49145  0.37245   0.12053  -5.894 3.77e-09 ***
## site06R -0.27350   0.76071  0.32654   0.09366  -2.920 0.003497 ** 
## site07R -0.55793   0.57239  0.30653   0.13255  -4.209 2.56e-05 ***
## site09R -0.18211   0.83351  0.34035   0.05269  -3.456 0.000548 ***
## site1   -0.04754   0.95358  0.28447   0.12715  -0.374 0.708512    
## site101 -0.24892   0.77964  0.29391   0.14116  -1.763 0.077819 .  
## site102 -0.18954   0.82734  0.29799   0.09288  -2.041 0.041281 *  
## site103  0.13423   1.14366  0.28632   0.07548   1.778 0.075324 .  
## site104 -0.16500   0.84790  0.29469   0.07366  -2.240 0.025094 *  
## site105 -0.06306   0.93889  0.29029   0.14070  -0.448 0.654014    
## site106 -0.42584   0.65322  0.28789   0.12037  -3.538 0.000403 ***
## site107  0.07045   1.07299  0.32733   0.14843   0.475 0.635063    
## site108  0.10534   1.11109  0.36947   0.18185   0.579 0.562404    
## site10R -0.24856   0.77992  0.38542   0.07079  -3.511 0.000446 ***
## site11R -0.33972   0.71197  0.30189   0.10312  -3.294 0.000987 ***
## site12R -0.16650   0.84662  0.33400   0.03958  -4.206 2.60e-05 ***
## site13R -0.51702   0.59630  0.30330   0.11929  -4.334 1.46e-05 ***
## site14R -0.60931   0.54373  0.64169   0.12406  -4.911 9.05e-07 ***
## site15R -0.17097   0.84285  0.33165   0.06002  -2.848 0.004394 ** 
## site16R -0.05003   0.95120  0.34541   0.05630  -0.889 0.374175    
## site17R -0.27263   0.76137  0.36269   0.08542  -3.192 0.001414 ** 
## site18R -0.50842   0.60145  0.32342   0.17325  -2.935 0.003339 ** 
## site19R  0.02922   1.02965  0.35000   0.01152   2.537 0.011193 *  
## site20R -0.43820   0.64520  0.34786   0.06160  -7.114 1.13e-12 ***
## site21R -0.62871   0.53328  0.32243   0.11748  -5.352 8.72e-08 ***
## site22R -0.33548   0.71499  0.30735   0.08213  -4.085 4.42e-05 ***
## site23R -0.37855   0.68485  0.32384   0.06814  -5.555 2.77e-08 ***
## site24R -0.26057   0.77061  0.33092   0.06113  -4.263 2.02e-05 ***
## site25R -0.25412   0.77560  0.32685   0.10257  -2.477 0.013232 *  
## site26R -0.60635   0.54534  0.34552   0.10949  -5.538 3.06e-08 ***
## site27R -0.25294   0.77652  0.76036   0.13914  -1.818 0.069085 .  
## site28R -0.51833   0.59551  0.43696   0.11709  -4.427 9.57e-06 ***
## site29R -0.45868   0.63212  0.43420   0.07187  -6.382 1.74e-10 ***
## site30R -0.60641   0.54531  0.33514   0.16363  -3.706 0.000211 ***
## site31R -0.40898   0.66433  0.44953   0.06549  -6.245 4.23e-10 ***
## site32R -0.52348   0.59246  0.36930   0.07090  -7.383 1.54e-13 ***
## site33R -0.53464   0.58588  0.33312   0.10677  -5.007 5.52e-07 ***
## site34R -0.24717   0.78101  0.31366   0.05295  -4.668 3.04e-06 ***
## site35R -0.15706   0.85465  0.33385   0.02707  -5.802 6.56e-09 ***
## site36R -0.50791   0.60176  0.34004   0.09347  -5.434 5.52e-08 ***
## site37R -0.47732   0.62045  0.33637   0.10662  -4.477 7.57e-06 ***
## site38R -0.47973   0.61895  0.32993   0.06720  -7.139 9.41e-13 ***
## site39R -0.30542   0.73682  0.36027   0.04093  -7.463 8.48e-14 ***
## site40R -0.45271   0.63590  0.35413   0.23145  -1.956 0.050464 .  
## site41R -0.42550   0.65344  0.36966   0.07719  -5.513 3.53e-08 ***
## site42R -0.19531   0.82257  0.36203   0.08142  -2.399 0.016446 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## PM         1.0935     0.9145    1.0409    1.1487
## dx_yr      1.1407     0.8767    0.9865    1.3190
## site02R    1.4770     0.6771    1.1791    1.8501
## site03R    0.7697     1.2992    0.7370    0.8039
## site04R    0.5814     1.7199    0.4466    0.7570
## site05R    0.4915     2.0348    0.3881    0.6224
## site06R    0.7607     1.3146    0.6331    0.9140
## site07R    0.5724     1.7470    0.4414    0.7422
## site09R    0.8335     1.1997    0.7517    0.9242
## site1      0.9536     1.0487    0.7432    1.2234
## site101    0.7796     1.2826    0.5912    1.0281
## site102    0.8273     1.2087    0.6896    0.9925
## site103    1.1437     0.8744    0.9864    1.3260
## site104    0.8479     1.1794    0.7339    0.9796
## site105    0.9389     1.0651    0.7126    1.2370
## site106    0.6532     1.5309    0.5159    0.8270
## site107    1.0730     0.9320    0.8021    1.4353
## site108    1.1111     0.9000    0.7780    1.5869
## site10R    0.7799     1.2822    0.6789    0.8960
## site11R    0.7120     1.4046    0.5817    0.8714
## site12R    0.8466     1.1812    0.7834    0.9149
## site13R    0.5963     1.6770    0.4720    0.7534
## site14R    0.5437     1.8392    0.4264    0.6934
## site15R    0.8428     1.1865    0.7493    0.9481
## site16R    0.9512     1.0513    0.8518    1.0622
## site17R    0.7614     1.3134    0.6440    0.9001
## site18R    0.6014     1.6627    0.4283    0.8446
## site19R    1.0296     0.9712    1.0067    1.0532
## site20R    0.6452     1.5499    0.5718    0.7280
## site21R    0.5333     1.8752    0.4236    0.6714
## site22R    0.7150     1.3986    0.6087    0.8399
## site23R    0.6849     1.4602    0.5992    0.7827
## site24R    0.7706     1.2977    0.6836    0.8687
## site25R    0.7756     1.2893    0.6344    0.9483
## site26R    0.5453     1.8337    0.4400    0.6759
## site27R    0.7765     1.2878    0.5912    1.0200
## site28R    0.5955     1.6792    0.4734    0.7491
## site29R    0.6321     1.5820    0.5491    0.7277
## site30R    0.5453     1.8338    0.3957    0.7515
## site31R    0.6643     1.5053    0.5843    0.7553
## site32R    0.5925     1.6879    0.5156    0.6808
## site33R    0.5859     1.7068    0.4753    0.7223
## site34R    0.7810     1.2804    0.7040    0.8664
## site35R    0.8546     1.1701    0.8105    0.9012
## site36R    0.6018     1.6618    0.5010    0.7227
## site37R    0.6204     1.6117    0.5034    0.7646
## site38R    0.6189     1.6156    0.5426    0.7061
## site39R    0.7368     1.3572    0.6800    0.7984
## site40R    0.6359     1.5726    0.4040    1.0009
## site41R    0.6534     1.5304    0.5617    0.7602
## site42R    0.8226     1.2157    0.7012    0.9649
## 
## Concordance= 0.598  (se = 0.042 )
## Likelihood ratio test= 393.1  on 51 df,   p=<2e-16
## Wald test            = 12.65  on 51 df,   p=1
## Score (logrank) test = 383.3  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ PM + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ PM + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = All_IPF, 
##     id = ID, cluster = cohort)
## 
##   n= 121241, number of events= 2709 
##    (2849 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## PM                  0.091837  1.096187  0.012965  0.025583   3.590 0.000331 ***
## dx_yr               0.129425  1.138174  0.008351  0.078994   1.638 0.101333    
## age_dx              0.003620  1.003627  0.002441  0.001862   1.945 0.051795 .  
## sexF               -0.147133  0.863179  0.044856  0.079342  -1.854 0.063680 .  
## dich_RaceNon-White  0.019630  1.019824  0.068805  0.065474   0.300 0.764319    
## smokeHxFormer       0.102733  1.108196  0.059312  0.024246   4.237 2.26e-05 ***
## smokeHxAlways      -0.049278  0.951916  0.138517  0.076317  -0.646 0.518467    
## smokeHxUnknown      0.313831  1.368658  0.153426  0.212565   1.476 0.139836    
## smokeHxEver         0.035221  1.035848  0.064338  0.005610   6.278 3.42e-10 ***
## disadv              0.038805  1.039568  0.070322  0.138014   0.281 0.778580    
## site02R             0.357983  1.430441  0.347672  0.120657   2.967 0.003008 ** 
## site03R            -0.249332  0.779322  0.357337  0.030023  -8.305  < 2e-16 ***
## site04R            -0.553654  0.574845  0.460054  0.140262  -3.947 7.90e-05 ***
## site05R            -0.727255  0.483234  0.383148  0.103508  -7.026 2.12e-12 ***
## site06R            -0.297273  0.742841  0.339386  0.106290  -2.797 0.005161 ** 
## site07R            -0.578704  0.560625  0.318322  0.126572  -4.572 4.83e-06 ***
## site09R            -0.240143  0.786515  0.352805  0.037444  -6.413 1.42e-10 ***
## site1              -0.134474  0.874176  0.300964  0.144984  -0.928 0.353663    
## site101            -0.316675  0.728568  0.309073  0.132713  -2.386 0.017025 *  
## site102            -0.267425  0.765348  0.313812  0.102359  -2.613 0.008985 ** 
## site103             0.050872  1.052189  0.301936  0.049705   1.023 0.306073    
## site104            -0.238952  0.787453  0.310440  0.052748  -4.530 5.90e-06 ***
## site105            -0.135963  0.872875  0.307531  0.126656  -1.073 0.283054    
## site106            -0.520068  0.594480  0.303967  0.107566  -4.835 1.33e-06 ***
## site107            -0.034024  0.966549  0.343073  0.126621  -0.269 0.788156    
## site108             0.019986  1.020187  0.381834  0.170829   0.117 0.906863    
## site10R            -0.319597  0.726441  0.396305  0.052045  -6.141 8.21e-10 ***
## site11R            -0.389621  0.677314  0.316340  0.088902  -4.383 1.17e-05 ***
## site12R            -0.196957  0.821226  0.345268  0.035094  -5.612 2.00e-08 ***
## site13R            -0.567420  0.566986  0.315020  0.120650  -4.703 2.56e-06 ***
## site14R            -0.633290  0.530842  0.647593  0.082096  -7.714 1.22e-14 ***
## site15R            -0.222817  0.800261  0.343927  0.052617  -4.235 2.29e-05 ***
## site16R            -0.064642  0.937403  0.355638  0.042196  -1.532 0.125534    
## site17R            -0.301495  0.739712  0.373681  0.066959  -4.503 6.71e-06 ***
## site18R            -0.552419  0.575556  0.335527  0.149705  -3.690 0.000224 ***
## site19R            -0.020529  0.979680  0.360824  0.023170  -0.886 0.375604    
## site20R            -0.450495  0.637312  0.359322  0.041368 -10.890  < 2e-16 ***
## site21R            -0.681061  0.506080  0.333935  0.110347  -6.172 6.74e-10 ***
## site22R            -0.362686  0.695805  0.319455  0.070735  -5.127 2.94e-07 ***
## site23R            -0.450334  0.637415  0.337731  0.061302  -7.346 2.04e-13 ***
## site24R            -0.282946  0.753561  0.342679  0.048680  -5.812 6.16e-09 ***
## site25R            -0.286930  0.750564  0.338773  0.114357  -2.509 0.012105 *  
## site26R            -0.657131  0.518336  0.356471  0.091405  -7.189 6.52e-13 ***
## site27R            -0.258245  0.772406  0.767858  0.138845  -1.860 0.062893 .  
## site28R            -0.536647  0.584705  0.445890  0.101618  -5.281 1.28e-07 ***
## site29R            -0.475255  0.621726  0.443492  0.090122  -5.273 1.34e-07 ***
## site30R            -0.689884  0.501634  0.347458  0.164160  -4.203 2.64e-05 ***
## site31R            -0.453042  0.635691  0.457746  0.058538  -7.739 1.00e-14 ***
## site32R            -0.603758  0.546753  0.389816  0.067813  -8.903  < 2e-16 ***
## site33R            -0.576704  0.561747  0.346291  0.083396  -6.915 4.67e-12 ***
## site34R            -0.270578  0.762938  0.326435  0.049271  -5.492 3.98e-08 ***
## site35R            -0.219282  0.803095  0.346941  0.018932 -11.583  < 2e-16 ***
## site36R            -0.529378  0.588971  0.351387  0.080397  -6.585 4.56e-11 ***
## site37R            -0.509113  0.601029  0.347352  0.103301  -4.928 8.29e-07 ***
## site38R            -0.498822  0.607245  0.341280  0.047607 -10.478  < 2e-16 ***
## site39R            -0.359653  0.697919  0.370984  0.035479 -10.137  < 2e-16 ***
## site40R            -0.523215  0.592612  0.364591  0.228534  -2.289 0.022054 *  
## site41R            -0.477021  0.620629  0.378999  0.073324  -6.506 7.74e-11 ***
## site42R            -0.235446  0.790218  0.372341  0.072252  -3.259 0.001119 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## PM                    1.0962     0.9123    1.0426    1.1526
## dx_yr                 1.1382     0.8786    0.9749    1.3288
## age_dx                1.0036     0.9964    1.0000    1.0073
## sexF                  0.8632     1.1585    0.7389    1.0084
## dich_RaceNon-White    1.0198     0.9806    0.8970    1.1595
## smokeHxFormer         1.1082     0.9024    1.0568    1.1621
## smokeHxAlways         0.9519     1.0505    0.8197    1.1055
## smokeHxUnknown        1.3687     0.7306    0.9023    2.0760
## smokeHxEver           1.0358     0.9654    1.0245    1.0473
## disadv                1.0396     0.9619    0.7932    1.3625
## site02R               1.4304     0.6991    1.1292    1.8121
## site03R               0.7793     1.2832    0.7348    0.8266
## site04R               0.5748     1.7396    0.4367    0.7567
## site05R               0.4832     2.0694    0.3945    0.5919
## site06R               0.7428     1.3462    0.6031    0.9149
## site07R               0.5606     1.7837    0.4375    0.7185
## site09R               0.7865     1.2714    0.7309    0.8464
## site1                 0.8742     1.1439    0.6579    1.1615
## site101               0.7286     1.3726    0.5617    0.9450
## site102               0.7653     1.3066    0.6262    0.9354
## site103               1.0522     0.9504    0.9545    1.1599
## site104               0.7875     1.2699    0.7101    0.8732
## site105               0.8729     1.1456    0.6810    1.1188
## site106               0.5945     1.6821    0.4815    0.7340
## site107               0.9665     1.0346    0.7541    1.2388
## site108               1.0202     0.9802    0.7299    1.4259
## site10R               0.7264     1.3766    0.6560    0.8045
## site11R               0.6773     1.4764    0.5690    0.8062
## site12R               0.8212     1.2177    0.7666    0.8797
## site13R               0.5670     1.7637    0.4476    0.7182
## site14R               0.5308     1.8838    0.4519    0.6235
## site15R               0.8003     1.2496    0.7218    0.8872
## site16R               0.9374     1.0668    0.8630    1.0182
## site17R               0.7397     1.3519    0.6487    0.8434
## site18R               0.5756     1.7375    0.4292    0.7718
## site19R               0.9797     1.0207    0.9362    1.0252
## site20R               0.6373     1.5691    0.5877    0.6911
## site21R               0.5061     1.9760    0.4077    0.6283
## site22R               0.6958     1.4372    0.6057    0.7993
## site23R               0.6374     1.5688    0.5653    0.7188
## site24R               0.7536     1.3270    0.6850    0.8290
## site25R               0.7506     1.3323    0.5999    0.9391
## site26R               0.5183     1.9292    0.4333    0.6200
## site27R               0.7724     1.2947    0.5884    1.0140
## site28R               0.5847     1.7103    0.4791    0.7136
## site29R               0.6217     1.6084    0.5211    0.7418
## site30R               0.5016     1.9935    0.3636    0.6920
## site31R               0.6357     1.5731    0.5668    0.7130
## site32R               0.5468     1.8290    0.4787    0.6245
## site33R               0.5617     1.7802    0.4770    0.6615
## site34R               0.7629     1.3107    0.6927    0.8403
## site35R               0.8031     1.2452    0.7738    0.8335
## site36R               0.5890     1.6979    0.5031    0.6895
## site37R               0.6010     1.6638    0.4909    0.7359
## site38R               0.6072     1.6468    0.5531    0.6666
## site39R               0.6979     1.4328    0.6510    0.7482
## site40R               0.5926     1.6874    0.3787    0.9275
## site41R               0.6206     1.6113    0.5375    0.7165
## site42R               0.7902     1.2655    0.6859    0.9104
## 
## Concordance= 0.603  (se = 0.038 )
## Likelihood ratio test= 426  on 59 df,   p=<2e-16
## Wald test            = 12.89  on 59 df,   p=1
## Score (logrank) test = 421.1  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

91.2 IPF-Only SO4

91.2.1 IPF-Only SO4 - Simmons Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ SO4 + dx_yr, data=Simm_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4 + dx_yr, data = Simm_IPF, 
##     id = ID)
## 
##   n= 31982, number of events= 695 
##    (446 observations deleted due to missingness)
## 
##          coef exp(coef) se(coef)     z Pr(>|z|)   
## SO4   0.09060   1.09483  0.06155 1.472  0.14099   
## dx_yr 0.04545   1.04650  0.01732 2.624  0.00868 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##       exp(coef) exp(-coef) lower .95 upper .95
## SO4       1.095     0.9134    0.9704     1.235
## dx_yr     1.046     0.9556    1.0116     1.083
## 
## Concordance= 0.52  (se = 0.013 )
## Likelihood ratio test= 9.8  on 2 df,   p=0.007
## Wald test            = 9.57  on 2 df,   p=0.008
## Score (logrank) test = 9.59  on 2 df,   p=0.008

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SO4 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv, data=Simm_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv, data = Simm_IPF, 
##     id = ID)
## 
##   n= 30488, number of events= 670 
##    (1940 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## SO4                 0.083150  1.086704  0.064814  1.283  0.19953    
## dx_yr               0.031789  1.032299  0.018531  1.715  0.08626 .  
## age_dx              0.001624  1.001625  0.004573  0.355  0.72252    
## sexF               -0.379354  0.684303  0.088721 -4.276  1.9e-05 ***
## dich_RaceNon-White  0.173916  1.189956  0.129043  1.348  0.17774    
## smokeHxFormer       0.081480  1.084892  0.091977  0.886  0.37569    
## smokeHxAlways      -0.402797  0.668447  0.266655 -1.511  0.13090    
## smokeHxUnknown      0.484776  1.623811  0.170538  2.843  0.00447 ** 
## disadv              0.373185  1.452352  0.134038  2.784  0.00537 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SO4                   1.0867     0.9202    0.9571    1.2339
## dx_yr                 1.0323     0.9687    0.9955    1.0705
## age_dx                1.0016     0.9984    0.9927    1.0106
## sexF                  0.6843     1.4613    0.5751    0.8143
## dich_RaceNon-White    1.1900     0.8404    0.9240    1.5324
## smokeHxFormer         1.0849     0.9218    0.9059    1.2992
## smokeHxAlways         0.6684     1.4960    0.3964    1.1273
## smokeHxUnknown        1.6238     0.6158    1.1624    2.2683
## disadv                1.4524     0.6885    1.1168    1.8887
## 
## Concordance= 0.59  (se = 0.012 )
## Likelihood ratio test= 53.29  on 9 df,   p=3e-08
## Wald test            = 53.64  on 9 df,   p=2e-08
## Score (logrank) test = 54.35  on 9 df,   p=2e-08

91.2.2 IPF-Only SO4 - PFF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ SO4 + dx_yr + site, data=PFF_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4 + dx_yr + 
##     site, data = PFF_IPF, id = ID)
## 
##   n= 54499, number of events= 1146 
##    (217 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef)      z Pr(>|z|)    
## SO4      0.29874   1.34816  0.11789  2.534   0.0113 *  
## dx_yr    0.15006   1.16190  0.02293  6.543 6.01e-11 ***
## site02R  0.51183   1.66834  0.34011  1.505   0.1324    
## site03R -0.29789   0.74239  0.34775 -0.857   0.3917    
## site04R -0.57325   0.56369  0.47156 -1.216   0.2241    
## site05R -0.86069   0.42287  0.42965 -2.003   0.0452 *  
## site06R -0.45893   0.63196  0.36799 -1.247   0.2124    
## site07R -0.67961   0.50681  0.33857 -2.007   0.0447 *  
## site08R       NA        NA  0.00000     NA       NA    
## site09R -0.29366   0.74553  0.35711 -0.822   0.4109    
## site10R -0.35920   0.69823  0.39396 -0.912   0.3619    
## site11R -0.47857   0.61967  0.34193 -1.400   0.1616    
## site12R -0.25982   0.77119  0.34126 -0.761   0.4464    
## site13R -0.69525   0.49895  0.34163 -2.035   0.0418 *  
## site14R -0.75645   0.46933  0.66132 -1.144   0.2527    
## site15R -0.32279   0.72413  0.35988 -0.897   0.3698    
## site16R -0.17098   0.84284  0.36529 -0.468   0.6397    
## site17R -0.39131   0.67617  0.39804 -0.983   0.3256    
## site18R -0.66029   0.51670  0.35747 -1.847   0.0647 .  
## site19R -0.05029   0.95096  0.35026 -0.144   0.8858    
## site20R -0.55355   0.57491  0.36740 -1.507   0.1319    
## site21R -0.71054   0.49138  0.36104 -1.968   0.0491 *  
## site22R -0.44350   0.64178  0.34101 -1.301   0.1934    
## site23R -0.52209   0.59328  0.35326 -1.478   0.1394    
## site24R -0.32669   0.72131  0.33912 -0.963   0.3354    
## site25R -0.10926   0.89650  0.32762 -0.333   0.7388    
## site26R -0.79588   0.45119  0.37927 -2.098   0.0359 *  
## site27R -0.41947   0.65739  0.77908 -0.538   0.5903    
## site28R -0.64660   0.52382  0.48181 -1.342   0.1796    
## site29R -0.57963   0.56010  0.44954 -1.289   0.1973    
## site30R -0.59736   0.55026  0.36543 -1.635   0.1021    
## site31R -0.65652   0.51866  0.46931 -1.399   0.1618    
## site32R -0.42321   0.65494  0.36991 -1.144   0.2526    
## site33R -0.74074   0.47676  0.36927 -2.006   0.0449 *  
## site34R -0.45032   0.63742  0.33614 -1.340   0.1803    
## site35R -0.30782   0.73505  0.34287 -0.898   0.3693    
## site36R -0.69292   0.50011  0.39089 -1.773   0.0763 .  
## site37R -0.62266   0.53652  0.38036 -1.637   0.1016    
## site38R -0.64687   0.52368  0.35936 -1.800   0.0719 .  
## site39R -0.24234   0.78479  0.36045 -0.672   0.5014    
## site40R -0.10430   0.90095  0.35966 -0.290   0.7718    
## site41R -0.51508   0.59745  0.38866 -1.325   0.1851    
## site42R -0.34951   0.70503  0.39792 -0.878   0.3798    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## SO4        1.3482     0.7418    1.0700    1.6986
## dx_yr      1.1619     0.8607    1.1108    1.2153
## site02R    1.6683     0.5994    0.8566    3.2493
## site03R    0.7424     1.3470    0.3755    1.4677
## site04R    0.5637     1.7740    0.2237    1.4205
## site05R    0.4229     2.3648    0.1822    0.9816
## site06R    0.6320     1.5824    0.3072    1.2999
## site07R    0.5068     1.9731    0.2610    0.9841
## site08R        NA         NA        NA        NA
## site09R    0.7455     1.3413    0.3702    1.5012
## site10R    0.6982     1.4322    0.3226    1.5113
## site11R    0.6197     1.6138    0.3170    1.2112
## site12R    0.7712     1.2967    0.3951    1.5054
## site13R    0.4990     2.0042    0.2554    0.9747
## site14R    0.4693     2.1307    0.1284    1.7155
## site15R    0.7241     1.3810    0.3577    1.4660
## site16R    0.8428     1.1865    0.4119    1.7246
## site17R    0.6762     1.4789    0.3099    1.4753
## site18R    0.5167     1.9354    0.2564    1.0412
## site19R    0.9510     1.0516    0.4786    1.8893
## site20R    0.5749     1.7394    0.2798    1.1812
## site21R    0.4914     2.0351    0.2422    0.9971
## site22R    0.6418     1.5582    0.3289    1.2522
## site23R    0.5933     1.6855    0.2969    1.1856
## site24R    0.7213     1.3864    0.3711    1.4021
## site25R    0.8965     1.1155    0.4717    1.7038
## site26R    0.4512     2.2164    0.2145    0.9488
## site27R    0.6574     1.5212    0.1428    3.0268
## site28R    0.5238     1.9090    0.2037    1.3468
## site29R    0.5601     1.7854    0.2321    1.3518
## site30R    0.5503     1.8173    0.2689    1.1262
## site31R    0.5187     1.9281    0.2067    1.3012
## site32R    0.6549     1.5268    0.3172    1.3523
## site33R    0.4768     2.0975    0.2312    0.9832
## site34R    0.6374     1.5688    0.3298    1.2318
## site35R    0.7350     1.3605    0.3754    1.4394
## site36R    0.5001     1.9995    0.2325    1.0760
## site37R    0.5365     1.8639    0.2546    1.1307
## site38R    0.5237     1.9096    0.2589    1.0591
## site39R    0.7848     1.2742    0.3872    1.5906
## site40R    0.9010     1.1099    0.4452    1.8233
## site41R    0.5975     1.6738    0.2789    1.2798
## site42R    0.7050     1.4184    0.3232    1.5379
## 
## Concordance= 0.59  (se = 0.009 )
## Likelihood ratio test= 116.8  on 42 df,   p=6e-09
## Wald test            = 115.6  on 42 df,   p=8e-09
## Score (logrank) test = 119.4  on 42 df,   p=2e-09

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SO4 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=PFF_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = PFF_IPF, 
##     id = ID)
## 
##   n= 53867, number of events= 1131 
##    (849 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)      z Pr(>|z|)    
## SO4                 0.3064731  1.3586250  0.1195415  2.564   0.0104 *  
## dx_yr               0.1500659  1.1619108  0.0234586  6.397 1.58e-10 ***
## age_dx             -0.0001734  0.9998266  0.0039376 -0.044   0.9649    
## sexM                0.0627253  1.0647343  0.0713136  0.880   0.3791    
## dich_RaceNon-White -0.0749020  0.9278344  0.1263413 -0.593   0.5533    
## smokeHxEver         0.0422050  1.0431083  0.0646237  0.653   0.5137    
## disadv             -0.0354140  0.9652058  0.1121920 -0.316   0.7523    
## site02R             0.4498402  1.5680615  0.3530122  1.274   0.2026    
## site03R            -0.3388478  0.7125909  0.3604354 -0.940   0.3472    
## site04R            -0.6439857  0.5251950  0.4832468 -1.333   0.1827    
## site05R            -0.9274047  0.3955790  0.4440119 -2.089   0.0367 *  
## site06R            -0.5194282  0.5948606  0.3858724 -1.346   0.1783    
## site07R            -0.7407730  0.4767452  0.3541161 -2.092   0.0364 *  
## site08R                    NA         NA  0.0000000     NA       NA    
## site09R            -0.3890314  0.6777130  0.3726994 -1.044   0.2966    
## site10R            -0.4351733  0.6471525  0.4085917 -1.065   0.2869    
## site11R            -0.5610823  0.5705912  0.3620088 -1.550   0.1212    
## site12R            -0.3211570  0.7253093  0.3562261 -0.902   0.3673    
## site13R            -0.7707461  0.4626677  0.3564345 -2.162   0.0306 *  
## site14R            -0.8345537  0.4340682  0.6698827 -1.246   0.2128    
## site15R            -0.3940048  0.6743508  0.3774856 -1.044   0.2966    
## site16R            -0.2290276  0.7953066  0.3775767 -0.607   0.5441    
## site17R            -0.4691302  0.6255461  0.4133164 -1.135   0.2564    
## site18R            -0.7184589  0.4875030  0.3750471 -1.916   0.0554 .  
## site19R            -0.1098669  0.8959534  0.3630151 -0.303   0.7622    
## site20R            -0.6226339  0.5365294  0.3826298 -1.627   0.1037    
## site21R            -0.7757126  0.4603756  0.3766455 -2.060   0.0394 *  
## site22R            -0.5140142  0.5980899  0.3564887 -1.442   0.1493    
## site23R            -0.6063407  0.5453428  0.3706927 -1.636   0.1019    
## site24R            -0.3958461  0.6731103  0.3537055 -1.119   0.2631    
## site25R            -0.1550409  0.8563802  0.3409623 -0.455   0.6493    
## site26R            -0.8523226  0.4264234  0.3944782 -2.161   0.0307 *  
## site27R            -0.5388276  0.5834318  0.7909146 -0.681   0.4957    
## site28R            -0.7030685  0.4950639  0.4958213 -1.418   0.1562    
## site29R            -0.6488605  0.5226410  0.4620503 -1.404   0.1602    
## site30R            -0.6698455  0.5117876  0.3797116 -1.764   0.0777 .  
## site31R            -0.7250954  0.4842783  0.4798081 -1.511   0.1307    
## site32R            -0.5100427  0.6004699  0.3926284 -1.299   0.1939    
## site33R            -0.7981224  0.4501734  0.3857167 -2.069   0.0385 *  
## site34R            -0.5115033  0.5995935  0.3528854 -1.449   0.1472    
## site35R            -0.3833053  0.6816048  0.3581951 -1.070   0.2846    
## site36R            -0.7504257  0.4721655  0.4077344 -1.840   0.0657 .  
## site37R            -0.6864738  0.5033478  0.3949042 -1.738   0.0822 .  
## site38R            -0.7037955  0.4947041  0.3745071 -1.879   0.0602 .  
## site39R            -0.3063334  0.7361411  0.3731401 -0.821   0.4117    
## site40R            -0.1477716  0.8626281  0.3711330 -0.398   0.6905    
## site41R            -0.5704288  0.5652830  0.4003813 -1.425   0.1542    
## site42R            -0.4327032  0.6487530  0.4114806 -1.052   0.2930    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SO4                   1.3586     0.7360    1.0748    1.7173
## dx_yr                 1.1619     0.8607    1.1097    1.2166
## age_dx                0.9998     1.0002    0.9921    1.0076
## sexM                  1.0647     0.9392    0.9258    1.2245
## dich_RaceNon-White    0.9278     1.0778    0.7243    1.1885
## smokeHxEver           1.0431     0.9587    0.9190    1.1840
## disadv                0.9652     1.0360    0.7747    1.2026
## site02R               1.5681     0.6377    0.7850    3.1322
## site03R               0.7126     1.4033    0.3516    1.4443
## site04R               0.5252     1.9041    0.2037    1.3541
## site05R               0.3956     2.5279    0.1657    0.9444
## site06R               0.5949     1.6811    0.2792    1.2673
## site07R               0.4767     2.0976    0.2382    0.9544
## site08R                   NA         NA        NA        NA
## site09R               0.6777     1.4756    0.3264    1.4070
## site10R               0.6472     1.5452    0.2905    1.4415
## site11R               0.5706     1.7526    0.2807    1.1600
## site12R               0.7253     1.3787    0.3608    1.4580
## site13R               0.4627     2.1614    0.2301    0.9304
## site14R               0.4341     2.3038    0.1168    1.6135
## site15R               0.6744     1.4829    0.3218    1.4132
## site16R               0.7953     1.2574    0.3794    1.6670
## site17R               0.6255     1.5986    0.2783    1.4063
## site18R               0.4875     2.0513    0.2337    1.0168
## site19R               0.8960     1.1161    0.4398    1.8251
## site20R               0.5365     1.8638    0.2535    1.1358
## site21R               0.4604     2.1721    0.2200    0.9632
## site22R               0.5981     1.6720    0.2974    1.2028
## site23R               0.5453     1.8337    0.2637    1.1277
## site24R               0.6731     1.4856    0.3365    1.3464
## site25R               0.8564     1.1677    0.4390    1.6707
## site26R               0.4264     2.3451    0.1968    0.9239
## site27R               0.5834     1.7140    0.1238    2.7493
## site28R               0.4951     2.0199    0.1873    1.3083
## site29R               0.5226     1.9134    0.2113    1.2927
## site30R               0.5118     1.9539    0.2432    1.0772
## site31R               0.4843     2.0649    0.1891    1.2402
## site32R               0.6005     1.6654    0.2782    1.2963
## site33R               0.4502     2.2214    0.2114    0.9587
## site34R               0.5996     1.6678    0.3002    1.1974
## site35R               0.6816     1.4671    0.3378    1.3754
## site36R               0.4722     2.1179    0.2123    1.0499
## site37R               0.5033     1.9867    0.2321    1.0915
## site38R               0.4947     2.0214    0.2374    1.0307
## site39R               0.7361     1.3584    0.3543    1.5296
## site40R               0.8626     1.1592    0.4168    1.7854
## site41R               0.5653     1.7690    0.2579    1.2390
## site42R               0.6488     1.5414    0.2896    1.4532
## 
## Concordance= 0.592  (se = 0.009 )
## Likelihood ratio test= 117.7  on 47 df,   p=5e-08
## Wald test            = 116.8  on 47 df,   p=7e-08
## Score (logrank) test = 120.6  on 47 df,   p=2e-08

91.2.3 IPF-Only SO4 - CARE-PF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ SO4 + dx_yr + site, data=CARE_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4 + dx_yr + 
##     site, data = CARE_IPF, id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef)      z Pr(>|z|)    
## SO4     -1.23596   0.29056  0.23269 -5.312 1.09e-07 ***
## dx_yr    0.73332   2.08198  0.03528 20.784  < 2e-16 ***
## site102 -0.60491   0.54612  0.21057 -2.873 0.004069 ** 
## site103 -0.33396   0.71608  0.19653 -1.699 0.089272 .  
## site104 -0.49248   0.61111  0.21087 -2.335 0.019521 *  
## site105 -0.11901   0.88780  0.12965 -0.918 0.358663    
## site106  0.35930   1.43233  0.12829  2.801 0.005098 ** 
## site107 -0.04064   0.96017  0.19854 -0.205 0.837800    
## site108 -0.98954   0.37175  0.28413 -3.483 0.000496 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## SO4        0.2906     3.4417    0.1841    0.4585
## dx_yr      2.0820     0.4803    1.9429    2.2310
## site102    0.5461     1.8311    0.3615    0.8251
## site103    0.7161     1.3965    0.4872    1.0526
## site104    0.6111     1.6364    0.4042    0.9239
## site105    0.8878     1.1264    0.6886    1.1447
## site106    1.4323     0.6982    1.1139    1.8418
## site107    0.9602     1.0415    0.6507    1.4169
## site108    0.3717     2.6900    0.2130    0.6488
## 
## Concordance= 0.776  (se = 0.01 )
## Likelihood ratio test= 1079  on 9 df,   p=<2e-16
## Wald test            = 691.5  on 9 df,   p=<2e-16
## Score (logrank) test = 586.5  on 9 df,   p=<2e-16

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SO4 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=CARE_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = CARE_IPF, 
##     id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## SO4                -1.286752  0.276166  0.238394 -5.398 6.75e-08 ***
## dx_yr               0.729942  2.074960  0.035557 20.529  < 2e-16 ***
## age_dx              0.012920  1.013004  0.004556  2.836 0.004570 ** 
## sexF               -0.046963  0.954123  0.077139 -0.609 0.542653    
## dich_RaceNon-White -0.123100  0.884176  0.110647 -1.113 0.265906    
## smokeHxFormer      -0.022266  0.977980  0.081559 -0.273 0.784853    
## smokeHxAlways       0.078855  1.082047  0.168784  0.467 0.640363    
## smokeHxUnknown      0.418192  1.519213  0.734668  0.569 0.569203    
## disadv              0.038522  1.039273  0.130494  0.295 0.767841    
## site102            -0.653165  0.520396  0.214710 -3.042 0.002349 ** 
## site103            -0.365703  0.693709  0.203983 -1.793 0.073003 .  
## site104            -0.524951  0.591584  0.215369 -2.437 0.014791 *  
## site105            -0.113263  0.892916  0.133202 -0.850 0.395154    
## site106             0.333095  1.395280  0.129390  2.574 0.010043 *  
## site107            -0.123085  0.884188  0.206526 -0.596 0.551188    
## site108            -1.027788  0.357797  0.290655 -3.536 0.000406 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SO4                   0.2762     3.6210    0.1731    0.4406
## dx_yr                 2.0750     0.4819    1.9353    2.2247
## age_dx                1.0130     0.9872    1.0040    1.0221
## sexF                  0.9541     1.0481    0.8202    1.1099
## dich_RaceNon-White    0.8842     1.1310    0.7118    1.0983
## smokeHxFormer         0.9780     1.0225    0.8335    1.1475
## smokeHxAlways         1.0820     0.9242    0.7773    1.5063
## smokeHxUnknown        1.5192     0.6582    0.3600    6.4116
## disadv                1.0393     0.9622    0.8047    1.3422
## site102               0.5204     1.9216    0.3416    0.7927
## site103               0.6937     1.4415    0.4651    1.0347
## site104               0.5916     1.6904    0.3879    0.9023
## site105               0.8929     1.1199    0.6877    1.1593
## site106               1.3953     0.7167    1.0827    1.7980
## site107               0.8842     1.1310    0.5899    1.3254
## site108               0.3578     2.7949    0.2024    0.6325
## 
## Concordance= 0.778  (se = 0.01 )
## Likelihood ratio test= 1091  on 16 df,   p=<2e-16
## Wald test            = 697.8  on 16 df,   p=<2e-16
## Score (logrank) test = 593.6  on 16 df,   p=<2e-16

91.2.4 IPF-Only SO4 - Combined Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ SO4 + dx_yr + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4 + dx_yr + 
##     site, data = All_IPF, id = ID, cluster = cohort)
## 
##   n= 123367, number of events= 2749 
##    (723 observations deleted due to missingness)
## 
##               coef  exp(coef)   se(coef)  robust se       z Pr(>|z|)    
## SO4      0.5110164  1.6669846  0.0377569  0.1279360   3.994 6.49e-05 ***
## dx_yr    0.2092500  1.2327532  0.0107065  0.0858800   2.437 0.014829 *  
## site02R  0.3812251  1.4640772  0.3348219  0.0862520   4.420 9.87e-06 ***
## site03R -0.3712861  0.6898466  0.3471583  0.0615363  -6.034 1.60e-09 ***
## site04R -0.8659004  0.4206726  0.4517212  0.1965069  -4.406 1.05e-05 ***
## site05R -1.2586311  0.2840426  0.3746006  0.1977626  -6.364 1.96e-10 ***
## site06R -0.7494408  0.4726308  0.3289288  0.0904956  -8.282  < 2e-16 ***
## site07R -0.9821490  0.3745054  0.3084016  0.2222340  -4.419 9.90e-06 ***
## site09R -0.5021582  0.6052230  0.3413533  0.0974425  -5.153 2.56e-07 ***
## site1   -0.7787003  0.4590022  0.2917903  0.1241201  -6.274 3.52e-10 ***
## site101 -0.6661744  0.5136699  0.2959348  0.2197325  -3.032 0.002431 ** 
## site102 -0.3388445  0.7125933  0.2980522  0.1039720  -3.259 0.001118 ** 
## site103 -0.0005837  0.9994164  0.2863842  0.0789955  -0.007 0.994104    
## site104 -0.3955723  0.6732946  0.2962273  0.1507913  -2.623 0.008708 ** 
## site105 -0.4105963  0.6632546  0.2916414  0.2037903  -2.015 0.043926 *  
## site106 -0.8462814  0.4290073  0.2899438  0.2093542  -4.042 5.29e-05 ***
## site107 -0.3112724  0.7325143  0.3287640  0.2092562  -1.488 0.136878    
## site108 -0.2267834  0.7970934  0.3702775  0.2129065  -1.065 0.286795    
## site10R -0.5245681  0.5918109  0.3860936  0.1281927  -4.092 4.28e-05 ***
## site11R -0.7874952  0.4549830  0.3038425  0.1900834  -4.143 3.43e-05 ***
## site12R -0.3891524  0.6776310  0.3345130  0.0500968  -7.768 7.97e-15 ***
## site13R -1.0086030  0.3647282  0.3059838  0.2181510  -4.623 3.77e-06 ***
## site14R -0.9941338  0.3700439  0.6424121  0.0457805 -21.715  < 2e-16 ***
## site15R -0.5780147  0.5610111  0.3332739  0.1279243  -4.518 6.23e-06 ***
## site16R -0.3702838  0.6905383  0.3462580  0.0665255  -5.566 2.61e-08 ***
## site17R -0.6885830  0.5022873  0.3637417  0.1513004  -4.551 5.34e-06 ***
## site18R -0.9765812  0.3765964  0.3256429  0.2708528  -3.606 0.000311 ***
## site19R -0.0973785  0.9072125  0.3499296  0.0233002  -4.179 2.92e-05 ***
## site20R -0.7847289  0.4562434  0.3490013  0.1174909  -6.679 2.40e-11 ***
## site21R -1.0384796  0.3539925  0.3235616  0.1861927  -5.577 2.44e-08 ***
## site22R -0.7158559  0.4887736  0.3085409  0.1307190  -5.476 4.34e-08 ***
## site23R -0.7858803  0.4557184  0.3254688  0.1417148  -5.546 2.93e-08 ***
## site24R -0.4604258  0.6310149  0.3313096  0.0448845 -10.258  < 2e-16 ***
## site25R -0.2224960  0.8005182  0.3252118  0.0831548  -2.676 0.007458 ** 
## site26R -1.1199960  0.3262811  0.3483931  0.2386110  -4.694 2.68e-06 ***
## site27R -0.7885837  0.4544880  0.7616707  0.2495674  -3.160 0.001579 ** 
## site28R -1.0307198  0.3567501  0.4383679  0.1840166  -5.601 2.13e-08 ***
## site29R -0.8190257  0.4408610  0.4352044  0.1392063  -5.884 4.02e-09 ***
## site30R -0.9102542  0.4024219  0.3342595  0.2279192  -3.994 6.50e-05 ***
## site31R -0.9261408  0.3960793  0.4513755  0.1716932  -5.394 6.88e-08 ***
## site32R -0.4981559  0.6076502  0.3686632  0.0676925  -7.359 1.85e-13 ***
## site33R -1.0704910  0.3428401  0.3362189  0.2328277  -4.598 4.27e-06 ***
## site34R -0.6904456  0.5013526  0.3156926  0.1471800  -4.691 2.72e-06 ***
## site35R -0.4596752  0.6314887  0.3346422  0.0903721  -5.086 3.65e-07 ***
## site36R -1.0548550  0.3482429  0.3429184  0.1792733  -5.884 4.00e-09 ***
## site37R -0.9660688  0.3805762  0.3385391  0.1918401  -5.036 4.76e-07 ***
## site38R -0.9170174  0.3997094  0.3319008  0.1435756  -6.387 1.69e-10 ***
## site39R -0.2879558  0.7497948  0.3600645  0.0295062  -9.759  < 2e-16 ***
## site40R -0.3403622  0.7115126  0.3440754  0.1769919  -1.923 0.054475 .  
## site41R -0.7468242  0.4738691  0.3704045  0.1329629  -5.617 1.95e-08 ***
## site42R -0.6534198  0.5202635  0.3637355  0.1597664  -4.090 4.32e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## SO4        1.6670     0.5999    1.2973    2.1421
## dx_yr      1.2328     0.8112    1.0418    1.4587
## site02R    1.4641     0.6830    1.2364    1.7337
## site03R    0.6898     1.4496    0.6115    0.7783
## site04R    0.4207     2.3771    0.2862    0.6183
## site05R    0.2840     3.5206    0.1928    0.4185
## site06R    0.4726     2.1158    0.3958    0.5644
## site07R    0.3745     2.6702    0.2423    0.5789
## site09R    0.6052     1.6523    0.5000    0.7326
## site1      0.4590     2.1786    0.3599    0.5854
## site101    0.5137     1.9468    0.3339    0.7902
## site102    0.7126     1.4033    0.5812    0.8737
## site103    0.9994     1.0006    0.8561    1.1668
## site104    0.6733     1.4852    0.5010    0.9048
## site105    0.6633     1.5077    0.4449    0.9889
## site106    0.4290     2.3310    0.2846    0.6466
## site107    0.7325     1.3652    0.4861    1.1039
## site108    0.7971     1.2546    0.5251    1.2099
## site10R    0.5918     1.6897    0.4603    0.7609
## site11R    0.4550     2.1979    0.3135    0.6604
## site12R    0.6776     1.4757    0.6143    0.7475
## site13R    0.3647     2.7418    0.2378    0.5593
## site14R    0.3700     2.7024    0.3383    0.4048
## site15R    0.5610     1.7825    0.4366    0.7209
## site16R    0.6905     1.4481    0.6061    0.7867
## site17R    0.5023     1.9909    0.3734    0.6757
## site18R    0.3766     2.6554    0.2215    0.6404
## site19R    0.9072     1.1023    0.8667    0.9496
## site20R    0.4562     2.1918    0.3624    0.5744
## site21R    0.3540     2.8249    0.2458    0.5099
## site22R    0.4888     2.0459    0.3783    0.6315
## site23R    0.4557     2.1943    0.3452    0.6016
## site24R    0.6310     1.5847    0.5779    0.6890
## site25R    0.8005     1.2492    0.6801    0.9422
## site26R    0.3263     3.0648    0.2044    0.5208
## site27R    0.4545     2.2003    0.2787    0.7412
## site28R    0.3568     2.8031    0.2487    0.5117
## site29R    0.4409     2.2683    0.3356    0.5792
## site30R    0.4024     2.4850    0.2574    0.6291
## site31R    0.3961     2.5247    0.2829    0.5545
## site32R    0.6077     1.6457    0.5321    0.6939
## site33R    0.3428     2.9168    0.2172    0.5411
## site34R    0.5014     1.9946    0.3757    0.6690
## site35R    0.6315     1.5836    0.5290    0.7539
## site36R    0.3482     2.8716    0.2451    0.4949
## site37R    0.3806     2.6276    0.2613    0.5543
## site38R    0.3997     2.5018    0.3017    0.5296
## site39R    0.7498     1.3337    0.7077    0.7944
## site40R    0.7115     1.4055    0.5030    1.0066
## site41R    0.4739     2.1103    0.3652    0.6149
## site42R    0.5203     1.9221    0.3804    0.7116
## 
## Concordance= 0.612  (se = 0.044 )
## Likelihood ratio test= 492.6  on 51 df,   p=<2e-16
## Wald test            = 116  on 51 df,   p=6e-07
## Score (logrank) test = 458.8  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SO4 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SO4 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All_IPF, 
##     id = ID, cluster = cohort)
## 
##   n= 121241, number of events= 2709 
##    (2849 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## SO4                 0.520926  1.683587  0.038357  0.126883   4.106 4.03e-05 ***
## dx_yr               0.208219  1.231483  0.011049  0.089361   2.330 0.019802 *  
## age_dx              0.002714  1.002717  0.002440  0.001988   1.365 0.172146    
## sexF               -0.142260  0.867396  0.044898  0.087104  -1.633 0.102423    
## dich_RaceNon-White  0.026394  1.026746  0.069038  0.077443   0.341 0.733237    
## smokeHxFormer       0.122602  1.130435  0.059501  0.012929   9.483  < 2e-16 ***
## smokeHxAlways       0.003729  1.003736  0.138752  0.117842   0.032 0.974754    
## smokeHxUnknown      0.337323  1.401192  0.153191  0.133651   2.524 0.011606 *  
## smokeHxEver         0.033698  1.034272  0.064183  0.003070  10.977  < 2e-16 ***
## disadv              0.050061  1.051335  0.070820  0.127129   0.394 0.693743    
## site02R             0.358416  1.431061  0.345503  0.098292   3.646 0.000266 ***
## site03R            -0.359633  0.697933  0.357528  0.074615  -4.820 1.44e-06 ***
## site04R            -0.876248  0.416342  0.460091  0.212618  -4.121 3.77e-05 ***
## site05R            -1.280350  0.277940  0.385123  0.193983  -6.600 4.10e-11 ***
## site06R            -0.783987  0.456582  0.341869  0.058329 -13.441  < 2e-16 ***
## site07R            -1.007042  0.365298  0.320141  0.217857  -4.623 3.79e-06 ***
## site09R            -0.558600  0.572009  0.353638  0.091666  -6.094 1.10e-09 ***
## site1              -0.890842  0.410310  0.309142  0.109898  -8.106 5.23e-16 ***
## site101            -0.751699  0.471565  0.311617  0.233845  -3.215 0.001307 ** 
## site102            -0.433928  0.647959  0.314090  0.137654  -3.152 0.001620 ** 
## site103            -0.104103  0.901132  0.302191  0.077670  -1.340 0.180139    
## site104            -0.489837  0.612726  0.311985  0.152522  -3.212 0.001320 ** 
## site105            -0.507254  0.602147  0.309478  0.217336  -2.334 0.019598 *  
## site106            -0.958363  0.383520  0.306594  0.222293  -4.311 1.62e-05 ***
## site107            -0.435567  0.646898  0.344920  0.221396  -1.967 0.049141 *  
## site108            -0.330708  0.718415  0.383011  0.225210  -1.468 0.141984    
## site10R            -0.592983  0.552676  0.396830  0.112457  -5.273 1.34e-07 ***
## site11R            -0.843849  0.430052  0.318160  0.179497  -4.701 2.59e-06 ***
## site12R            -0.422370  0.655492  0.345748  0.040390 -10.457  < 2e-16 ***
## site13R            -1.060599  0.346248  0.317541  0.218003  -4.865 1.14e-06 ***
## site14R            -1.018665  0.361077  0.648296  0.041257 -24.691  < 2e-16 ***
## site15R            -0.635002  0.529934  0.345472  0.106049  -5.988 2.13e-09 ***
## site16R            -0.390440  0.676759  0.356497  0.052564  -7.428 1.10e-13 ***
## site17R            -0.725185  0.484235  0.374751  0.144867  -5.006 5.56e-07 ***
## site18R            -1.025649  0.358564  0.337674  0.246077  -4.168 3.07e-05 ***
## site19R            -0.144486  0.865467  0.360587  0.014129 -10.226  < 2e-16 ***
## site20R            -0.800300  0.449194  0.360432  0.119093  -6.720 1.82e-11 ***
## site21R            -1.091467  0.335724  0.334911  0.174087  -6.270 3.62e-10 ***
## site22R            -0.743512  0.475441  0.320519  0.134476  -5.529 3.22e-08 ***
## site23R            -0.857260  0.424323  0.339231  0.120987  -7.086 1.39e-12 ***
## site24R            -0.480789  0.618295  0.342961  0.032167 -14.947  < 2e-16 ***
## site25R            -0.252844  0.776589  0.336897  0.101385  -2.494 0.012635 *  
## site26R            -1.177002  0.308201  0.359258  0.215928  -5.451 5.01e-08 ***
## site27R            -0.806733  0.446314  0.769289  0.245361  -3.288 0.001009 ** 
## site28R            -1.053512  0.348711  0.447268  0.170000  -6.197 5.75e-10 ***
## site29R            -0.833342  0.434595  0.444429  0.175186  -4.757 1.97e-06 ***
## site30R            -0.986645  0.372826  0.346247  0.223610  -4.412 1.02e-05 ***
## site31R            -0.974724  0.377297  0.459506  0.164924  -5.910 3.42e-09 ***
## site32R            -0.574440  0.563020  0.388897  0.069742  -8.237  < 2e-16 ***
## site33R            -1.110989  0.329233  0.349163  0.205275  -5.412 6.23e-08 ***
## site34R            -0.715291  0.489050  0.328338  0.150969  -4.738 2.16e-06 ***
## site35R            -0.518945  0.595148  0.347588  0.081002  -6.407 1.49e-10 ***
## site36R            -1.080404  0.339458  0.354157  0.170323  -6.343 2.25e-10 ***
## site37R            -0.999994  0.367881  0.349418  0.196010  -5.102 3.37e-07 ***
## site38R            -0.944552  0.388854  0.343268  0.120605  -7.832 4.81e-15 ***
## site39R            -0.330590  0.718500  0.370684  0.020696 -15.973  < 2e-16 ***
## site40R            -0.400477  0.670001  0.354043  0.152032  -2.634 0.008435 ** 
## site41R            -0.800329  0.449181  0.379676  0.121612  -6.581 4.67e-11 ***
## site42R            -0.695337  0.498906  0.373927  0.159468  -4.360 1.30e-05 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SO4                   1.6836     0.5940    1.3129    2.1589
## dx_yr                 1.2315     0.8120    1.0336    1.4672
## age_dx                1.0027     0.9973    0.9988    1.0066
## sexF                  0.8674     1.1529    0.7313    1.0289
## dich_RaceNon-White    1.0267     0.9740    0.8822    1.1950
## smokeHxFormer         1.1304     0.8846    1.1021    1.1594
## smokeHxAlways         1.0037     0.9963    0.7967    1.2645
## smokeHxUnknown        1.4012     0.7137    1.0783    1.8208
## smokeHxEver           1.0343     0.9669    1.0281    1.0405
## disadv                1.0513     0.9512    0.8195    1.3488
## site02R               1.4311     0.6988    1.1803    1.7351
## site03R               0.6979     1.4328    0.6030    0.8078
## site04R               0.4163     2.4019    0.2745    0.6316
## site05R               0.2779     3.5979    0.1900    0.4065
## site06R               0.4566     2.1902    0.4073    0.5119
## site07R               0.3653     2.7375    0.2383    0.5599
## site09R               0.5720     1.7482    0.4779    0.6846
## site1                 0.4103     2.4372    0.3308    0.5089
## site101               0.4716     2.1206    0.2982    0.7457
## site102               0.6480     1.5433    0.4947    0.8486
## site103               0.9011     1.1097    0.7739    1.0493
## site104               0.6127     1.6321    0.4544    0.8262
## site105               0.6021     1.6607    0.3933    0.9219
## site106               0.3835     2.6074    0.2481    0.5929
## site107               0.6469     1.5458    0.4192    0.9984
## site108               0.7184     1.3920    0.4620    1.1171
## site10R               0.5527     1.8094    0.4434    0.6890
## site11R               0.4301     2.3253    0.3025    0.6114
## site12R               0.6555     1.5256    0.6056    0.7095
## site13R               0.3462     2.8881    0.2259    0.5308
## site14R               0.3611     2.7695    0.3330    0.3915
## site15R               0.5299     1.8870    0.4305    0.6524
## site16R               0.6768     1.4776    0.6105    0.7502
## site17R               0.4842     2.0651    0.3645    0.6432
## site18R               0.3586     2.7889    0.2214    0.5808
## site19R               0.8655     1.1554    0.8418    0.8898
## site20R               0.4492     2.2262    0.3557    0.5673
## site21R               0.3357     2.9786    0.2387    0.4722
## site22R               0.4754     2.1033    0.3653    0.6188
## site23R               0.4243     2.3567    0.3347    0.5379
## site24R               0.6183     1.6173    0.5805    0.6585
## site25R               0.7766     1.2877    0.6366    0.9473
## site26R               0.3082     3.2446    0.2019    0.4706
## site27R               0.4463     2.2406    0.2759    0.7219
## site28R               0.3487     2.8677    0.2499    0.4866
## site29R               0.4346     2.3010    0.3083    0.6126
## site30R               0.3728     2.6822    0.2405    0.5779
## site31R               0.3773     2.6504    0.2731    0.5213
## site32R               0.5630     1.7761    0.4911    0.6455
## site33R               0.3292     3.0374    0.2202    0.4923
## site34R               0.4890     2.0448    0.3638    0.6574
## site35R               0.5951     1.6803    0.5078    0.6975
## site36R               0.3395     2.9459    0.2431    0.4740
## site37R               0.3679     2.7183    0.2505    0.5402
## site38R               0.3889     2.5717    0.3070    0.4925
## site39R               0.7185     1.3918    0.6899    0.7482
## site40R               0.6700     1.4925    0.4974    0.9026
## site41R               0.4492     2.2263    0.3539    0.5701
## site42R               0.4989     2.0044    0.3650    0.6820
## 
## Concordance= 0.617  (se = 0.041 )
## Likelihood ratio test= 525.5  on 59 df,   p=<2e-16
## Wald test            = 130.7  on 59 df,   p=2e-07
## Score (logrank) test = 494.4  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

91.3 IPF-Only NO3

91.3.1 IPF-Only NO3 - Simmons Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ NO3 + dx_yr, data=Simm_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3 + dx_yr, data = Simm_IPF, 
##     id = ID)
## 
##   n= 31982, number of events= 695 
##    (446 observations deleted due to missingness)
## 
##           coef exp(coef) se(coef)     z Pr(>|z|)   
## NO3   0.078119  1.081251 0.120972 0.646  0.51844   
## dx_yr 0.025192  1.025512 0.008946 2.816  0.00486 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##       exp(coef) exp(-coef) lower .95 upper .95
## NO3       1.081     0.9249     0.853     1.371
## dx_yr     1.026     0.9751     1.008     1.044
## 
## Concordance= 0.523  (se = 0.013 )
## Likelihood ratio test= 8  on 2 df,   p=0.02
## Wald test            = 7.97  on 2 df,   p=0.02
## Score (logrank) test = 7.99  on 2 df,   p=0.02

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NO3 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv, data=Simm_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv, data = Simm_IPF, 
##     id = ID)
## 
##   n= 30488, number of events= 670 
##    (1940 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## NO3                 0.065691  1.067897  0.126750  0.518  0.60427    
## dx_yr               0.012666  1.012746  0.009522  1.330  0.18348    
## age_dx              0.002397  1.002400  0.004560  0.526  0.59908    
## sexF               -0.377364  0.685666  0.089126 -4.234  2.3e-05 ***
## dich_RaceNon-White  0.152740  1.165022  0.128781  1.186  0.23561    
## smokeHxFormer       0.087396  1.091329  0.091830  0.952  0.34124    
## smokeHxAlways      -0.407825  0.665095  0.266536 -1.530  0.12599    
## smokeHxUnknown      0.508275  1.662421  0.169685  2.995  0.00274 ** 
## disadv              0.381580  1.464597  0.134657  2.834  0.00460 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NO3                   1.0679     0.9364    0.8330    1.3690
## dx_yr                 1.0127     0.9874    0.9940    1.0318
## age_dx                1.0024     0.9976    0.9935    1.0114
## sexF                  0.6857     1.4584    0.5758    0.8165
## dich_RaceNon-White    1.1650     0.8584    0.9051    1.4995
## smokeHxFormer         1.0913     0.9163    0.9116    1.3065
## smokeHxAlways         0.6651     1.5035    0.3945    1.1214
## smokeHxUnknown        1.6624     0.6015    1.1921    2.3183
## disadv                1.4646     0.6828    1.1249    1.9069
## 
## Concordance= 0.591  (se = 0.012 )
## Likelihood ratio test= 51.89  on 9 df,   p=5e-08
## Wald test            = 52.36  on 9 df,   p=4e-08
## Score (logrank) test = 53.1  on 9 df,   p=3e-08

91.3.2 IPF-Only NO3 - PFF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ NO3 + dx_yr + site, data=PFF_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3 + dx_yr + 
##     site, data = PFF_IPF, id = ID)
## 
##   n= 54499, number of events= 1146 
##    (217 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef)      z Pr(>|z|)    
## NO3     -0.10347   0.90170  0.09648 -1.072    0.284    
## dx_yr    0.10001   1.10518  0.01409  7.096 1.28e-12 ***
## site02R  0.75122   2.11958  0.34621  2.170    0.030 *  
## site03R -0.21561   0.80605  0.34813 -0.619    0.536    
## site04R -0.06236   0.93954  0.46930 -0.133    0.894    
## site05R -0.21424   0.80715  0.37696 -0.568    0.570    
## site06R  0.00428   1.00429  0.32606  0.013    0.990    
## site07R -0.19408   0.82359  0.32121 -0.604    0.546    
## site08R       NA        NA  0.00000     NA       NA    
## site09R  0.03011   1.03057  0.34203  0.088    0.930    
## site10R -0.15118   0.85969  0.38536 -0.392    0.695    
## site11R -0.05390   0.94753  0.29955 -0.180    0.857    
## site12R  0.00305   1.00306  0.34261  0.009    0.993    
## site13R -0.23298   0.79217  0.30593 -0.762    0.446    
## site14R -0.26079   0.77044  0.64636 -0.403    0.687    
## site15R  0.03762   1.03834  0.33050  0.114    0.909    
## site16R  0.20490   1.22740  0.34915  0.587    0.557    
## site17R  0.13644   1.14618  0.37027  0.368    0.713    
## site18R -0.25622   0.77398  0.32260 -0.794    0.427    
## site19R -0.02996   0.97049  0.35044 -0.085    0.932    
## site20R -0.16763   0.84567  0.35413 -0.473    0.636    
## site21R -0.26543   0.76687  0.31967 -0.830    0.406    
## site22R  0.03373   1.03430  0.31570  0.107    0.915    
## site23R -0.11986   0.88704  0.32376 -0.370    0.711    
## site24R -0.03264   0.96788  0.34420 -0.095    0.924    
## site25R  0.08598   1.08978  0.33518  0.257    0.798    
## site26R -0.38737   0.67884  0.34465 -1.124    0.261    
## site27R  0.10566   1.11144  0.76372  0.138    0.890    
## site28R -0.01218   0.98789  0.44320 -0.027    0.978    
## site29R -0.23190   0.79302  0.43549 -0.533    0.594    
## site30R -0.04987   0.95135  0.35833 -0.139    0.889    
## site31R -0.29876   0.74174  0.44961 -0.664    0.506    
## site32R -0.32873   0.71984  0.36871 -0.892    0.373    
## site33R -0.31342   0.73095  0.33261 -0.942    0.346    
## site34R -0.09133   0.91272  0.31636 -0.289    0.773    
## site35R -0.09434   0.90997  0.33396 -0.282    0.778    
## site36R -0.08052   0.92263  0.35260 -0.228    0.819    
## site37R -0.05496   0.94652  0.34745 -0.158    0.874    
## site38R -0.26833   0.76465  0.32907 -0.815    0.415    
## site39R -0.13661   0.87231  0.36429 -0.375    0.708    
## site40R  0.38370   1.46770  0.39266  0.977    0.328    
## site41R -0.12718   0.88058  0.37427 -0.340    0.734    
## site42R  0.15895   1.17228  0.36632  0.434    0.664    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## NO3        0.9017     1.1090    0.7463     1.089
## dx_yr      1.1052     0.9048    1.0751     1.136
## site02R    2.1196     0.4718    1.0754     4.178
## site03R    0.8060     1.2406    0.4074     1.595
## site04R    0.9395     1.0643    0.3745     2.357
## site05R    0.8072     1.2389    0.3856     1.690
## site06R    1.0043     0.9957    0.5300     1.903
## site07R    0.8236     1.2142    0.4388     1.546
## site08R        NA         NA        NA        NA
## site09R    1.0306     0.9703    0.5272     2.015
## site10R    0.8597     1.1632    0.4039     1.830
## site11R    0.9475     1.0554    0.5268     1.704
## site12R    1.0031     0.9970    0.5125     1.963
## site13R    0.7922     1.2624    0.4349     1.443
## site14R    0.7704     1.2980    0.2170     2.735
## site15R    1.0383     0.9631    0.5433     1.985
## site16R    1.2274     0.8147    0.6191     2.433
## site17R    1.1462     0.8725    0.5547     2.368
## site18R    0.7740     1.2920    0.4113     1.457
## site19R    0.9705     1.0304    0.4883     1.929
## site20R    0.8457     1.1825    0.4224     1.693
## site21R    0.7669     1.3040    0.4098     1.435
## site22R    1.0343     0.9668    0.5571     1.920
## site23R    0.8870     1.1273    0.4703     1.673
## site24R    0.9679     1.0332    0.4930     1.900
## site25R    1.0898     0.9176    0.5650     2.102
## site26R    0.6788     1.4731    0.3455     1.334
## site27R    1.1114     0.8997    0.2488     4.966
## site28R    0.9879     1.0123    0.4144     2.355
## site29R    0.7930     1.2610    0.3377     1.862
## site30R    0.9514     1.0511    0.4713     1.920
## site31R    0.7417     1.3482    0.3073     1.790
## site32R    0.7198     1.3892    0.3495     1.483
## site33R    0.7309     1.3681    0.3809     1.403
## site34R    0.9127     1.0956    0.4910     1.697
## site35R    0.9100     1.0989    0.4729     1.751
## site36R    0.9226     1.0839    0.4623     1.841
## site37R    0.9465     1.0565    0.4790     1.870
## site38R    0.7647     1.3078    0.4012     1.457
## site39R    0.8723     1.1464    0.4272     1.781
## site40R    1.4677     0.6813    0.6798     3.169
## site41R    0.8806     1.1356    0.4229     1.834
## site42R    1.1723     0.8530    0.5718     2.403
## 
## Concordance= 0.589  (se = 0.009 )
## Likelihood ratio test= 111.7  on 42 df,   p=3e-08
## Wald test            = 110  on 42 df,   p=5e-08
## Score (logrank) test = 114.3  on 42 df,   p=1e-08

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NO3 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=PFF_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = PFF_IPF, 
##     id = ID)
## 
##   n= 53867, number of events= 1131 
##    (849 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)      z Pr(>|z|)    
## NO3                -0.0996736  0.9051328  0.0977188 -1.020   0.3077    
## dx_yr               0.0986732  1.1037056  0.0145693  6.773 1.26e-11 ***
## age_dx             -0.0001214  0.9998786  0.0039361 -0.031   0.9754    
## sexM                0.0638078  1.0658875  0.0713192  0.895   0.3710    
## dich_RaceNon-White -0.0519724  0.9493550  0.1259445 -0.413   0.6799    
## smokeHxEver         0.0445620  1.0455698  0.0646169  0.690   0.4904    
## disadv             -0.0544620  0.9469945  0.1122059 -0.485   0.6274    
## site02R             0.7071011  2.0281035  0.3589002  1.970   0.0488 *  
## site03R            -0.2374812  0.7886117  0.3604155 -0.659   0.5100    
## site04R            -0.1124773  0.8936177  0.4804784 -0.234   0.8149    
## site05R            -0.2524409  0.7769022  0.3894613 -0.648   0.5169    
## site06R            -0.0238829  0.9764001  0.3415139 -0.070   0.9442    
## site07R            -0.2316936  0.7931891  0.3357447 -0.690   0.4901    
## site08R                    NA         NA  0.0000000     NA       NA    
## site09R            -0.0441533  0.9568073  0.3563955 -0.124   0.9014    
## site10R            -0.2001765  0.8185862  0.3984932 -0.502   0.6154    
## site11R            -0.1005138  0.9043727  0.3170201 -0.317   0.7512    
## site12R            -0.0352559  0.9653583  0.3570301 -0.099   0.9213    
## site13R            -0.2837648  0.7529437  0.3191352 -0.889   0.3739    
## site14R            -0.3154306  0.7294747  0.6538639 -0.482   0.6295    
## site15R            -0.0010874  0.9989132  0.3455663 -0.003   0.9975    
## site16R             0.1658084  1.1803469  0.3606762  0.460   0.6457    
## site17R             0.0857695  1.0895551  0.3840186  0.223   0.8233    
## site18R            -0.2810226  0.7550113  0.3376581 -0.832   0.4053    
## site19R            -0.0683353  0.9339473  0.3626204 -0.188   0.8505    
## site20R            -0.2127411  0.8083654  0.3682848 -0.578   0.5635    
## site21R            -0.3003307  0.7405732  0.3329195 -0.902   0.3670    
## site22R            -0.0136384  0.9864541  0.3300969 -0.041   0.9670    
## site23R            -0.1750541  0.8394116  0.3393329 -0.516   0.6059    
## site24R            -0.0834380  0.9199481  0.3584811 -0.233   0.8160    
## site25R             0.0582457  1.0599754  0.3484858  0.167   0.8673    
## site26R            -0.4129137  0.6617194  0.3574071 -1.155   0.2480    
## site27R             0.0072336  1.0072599  0.7748598  0.009   0.9926    
## site28R            -0.0362095  0.9644383  0.4551403 -0.080   0.9366    
## site29R            -0.2788161  0.7566790  0.4468028 -0.624   0.5326    
## site30R            -0.1004341  0.9044447  0.3723484 -0.270   0.7874    
## site31R            -0.3421022  0.7102756  0.4584973 -0.746   0.4556    
## site32R            -0.3742243  0.6878226  0.3905748 -0.958   0.3380    
## site33R            -0.3481023  0.7060266  0.3479638 -1.000   0.3171    
## site34R            -0.1274019  0.8803797  0.3317365 -0.384   0.7009    
## site35R            -0.1466239  0.8636187  0.3481678 -0.421   0.6737    
## site36R            -0.1066625  0.8988290  0.3675072 -0.290   0.7716    
## site37R            -0.0934345  0.9107977  0.3605854 -0.259   0.7955    
## site38R            -0.2970943  0.7429739  0.3423768 -0.868   0.3855    
## site39R            -0.1815916  0.8339419  0.3769172 -0.482   0.6300    
## site40R             0.3515743  1.4213033  0.4044017  0.869   0.3846    
## site41R            -0.1603955  0.8518068  0.3847990 -0.417   0.6768    
## site42R             0.0993722  1.1044773  0.3784940  0.263   0.7929    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NO3                   0.9051     1.1048    0.7474     1.096
## dx_yr                 1.1037     0.9060    1.0726     1.136
## age_dx                0.9999     1.0001    0.9922     1.008
## sexM                  1.0659     0.9382    0.9268     1.226
## dich_RaceNon-White    0.9494     1.0533    0.7417     1.215
## smokeHxEver           1.0456     0.9564    0.9212     1.187
## disadv                0.9470     1.0560    0.7600     1.180
## site02R               2.0281     0.4931    1.0037     4.098
## site03R               0.7886     1.2681    0.3891     1.598
## site04R               0.8936     1.1190    0.3485     2.292
## site05R               0.7769     1.2872    0.3621     1.667
## site06R               0.9764     1.0242    0.5000     1.907
## site07R               0.7932     1.2607    0.4108     1.532
## site08R                   NA         NA        NA        NA
## site09R               0.9568     1.0451    0.4758     1.924
## site10R               0.8186     1.2216    0.3749     1.788
## site11R               0.9044     1.1057    0.4858     1.683
## site12R               0.9654     1.0359    0.4795     1.944
## site13R               0.7529     1.3281    0.4028     1.407
## site14R               0.7295     1.3708    0.2025     2.628
## site15R               0.9989     1.0011    0.5074     1.966
## site16R               1.1803     0.8472    0.5821     2.393
## site17R               1.0896     0.9178    0.5133     2.313
## site18R               0.7550     1.3245    0.3895     1.463
## site19R               0.9339     1.0707    0.4588     1.901
## site20R               0.8084     1.2371    0.3928     1.664
## site21R               0.7406     1.3503    0.3856     1.422
## site22R               0.9865     1.0137    0.5165     1.884
## site23R               0.8394     1.1913    0.4317     1.632
## site24R               0.9199     1.0870    0.4556     1.857
## site25R               1.0600     0.9434    0.5354     2.099
## site26R               0.6617     1.5112    0.3284     1.333
## site27R               1.0073     0.9928    0.2206     4.599
## site28R               0.9644     1.0369    0.3952     2.353
## site29R               0.7567     1.3216    0.3152     1.816
## site30R               0.9044     1.1057    0.4360     1.876
## site31R               0.7103     1.4079    0.2892     1.745
## site32R               0.6878     1.4539    0.3199     1.479
## site33R               0.7060     1.4164    0.3570     1.396
## site34R               0.8804     1.1359    0.4595     1.687
## site35R               0.8636     1.1579    0.4365     1.709
## site36R               0.8988     1.1126    0.4374     1.847
## site37R               0.9108     1.0979    0.4493     1.847
## site38R               0.7430     1.3459    0.3798     1.453
## site39R               0.8339     1.1991    0.3984     1.746
## site40R               1.4213     0.7036    0.6434     3.140
## site41R               0.8518     1.1740    0.4007     1.811
## site42R               1.1045     0.9054    0.5260     2.319
## 
## Concordance= 0.591  (se = 0.009 )
## Likelihood ratio test= 112.2  on 47 df,   p=3e-07
## Wald test            = 110.8  on 47 df,   p=4e-07
## Score (logrank) test = 115.1  on 47 df,   p=1e-07

91.3.3 IPF-Only NO3 - CARE-PF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ NO3 + dx_yr + site, data=CARE_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3 + dx_yr + 
##     site, data = CARE_IPF, id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef)      z Pr(>|z|)    
## NO3     -0.270478  0.763014  0.196328 -1.378   0.1683    
## dx_yr    0.808432  2.244386  0.032614 24.788   <2e-16 ***
## site102  0.118986  1.126354  0.164897  0.722   0.4706    
## site103  0.372193  1.450913  0.160649  2.317   0.0205 *  
## site104  0.227337  1.255252  0.174687  1.301   0.1931    
## site105  0.006732  1.006755  0.133282  0.051   0.9597    
## site106  0.167708  1.182592  0.123642  1.356   0.1750    
## site107  0.052914  1.054339  0.201123  0.263   0.7925    
## site108 -0.449046  0.638236  0.266256 -1.687   0.0917 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## NO3        0.7630     1.3106    0.5193     1.121
## dx_yr      2.2444     0.4456    2.1054     2.393
## site102    1.1264     0.8878    0.8153     1.556
## site103    1.4509     0.6892    1.0590     1.988
## site104    1.2553     0.7967    0.8913     1.768
## site105    1.0068     0.9933    0.7753     1.307
## site106    1.1826     0.8456    0.9281     1.507
## site107    1.0543     0.9485    0.7109     1.564
## site108    0.6382     1.5668    0.3787     1.076
## 
## Concordance= 0.774  (se = 0.01 )
## Likelihood ratio test= 1054  on 9 df,   p=<2e-16
## Wald test            = 666.7  on 9 df,   p=<2e-16
## Score (logrank) test = 586.4  on 9 df,   p=<2e-16

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NO3 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=CARE_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = CARE_IPF, 
##     id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)      z Pr(>|z|)    
## NO3                -0.3037208  0.7380669  0.1996932 -1.521  0.12828    
## dx_yr               0.8073192  2.2418900  0.0327554 24.647  < 2e-16 ***
## age_dx              0.0121849  1.0122595  0.0045119  2.701  0.00692 ** 
## sexF               -0.0441648  0.9567963  0.0770348 -0.573  0.56644    
## dich_RaceNon-White -0.1468558  0.8634185  0.1103880 -1.330  0.18340    
## smokeHxFormer      -0.0002003  0.9997998  0.0815445 -0.002  0.99804    
## smokeHxAlways       0.0518455  1.0532130  0.1686644  0.307  0.75855    
## smokeHxUnknown      0.4203794  1.5225390  0.7347411  0.572  0.56722    
## disadv              0.0006266  1.0006268  0.1301703  0.005  0.99616    
## site102             0.0888252  1.0928896  0.1668224  0.532  0.59441    
## site103             0.3700427  1.4477964  0.1644577  2.250  0.02444 *  
## site104             0.2179053  1.2434693  0.1769330  1.232  0.21811    
## site105             0.0174679  1.0176214  0.1368009  0.128  0.89840    
## site106             0.1401743  1.1504743  0.1247604  1.124  0.26120    
## site107            -0.0165668  0.9835697  0.2090597 -0.079  0.93684    
## site108            -0.4468570  0.6396354  0.2694624 -1.658  0.09725 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NO3                   0.7381     1.3549    0.4990     1.092
## dx_yr                 2.2419     0.4461    2.1025     2.391
## age_dx                1.0123     0.9879    1.0033     1.021
## sexF                  0.9568     1.0452    0.8227     1.113
## dich_RaceNon-White    0.8634     1.1582    0.6954     1.072
## smokeHxFormer         0.9998     1.0002    0.8521     1.173
## smokeHxAlways         1.0532     0.9495    0.7567     1.466
## smokeHxUnknown        1.5225     0.6568    0.3607     6.427
## disadv                1.0006     0.9994    0.7753     1.291
## site102               1.0929     0.9150    0.7881     1.516
## site103               1.4478     0.6907    1.0489     1.998
## site104               1.2435     0.8042    0.8791     1.759
## site105               1.0176     0.9827    0.7783     1.331
## site106               1.1505     0.8692    0.9009     1.469
## site107               0.9836     1.0167    0.6529     1.482
## site108               0.6396     1.5634    0.3772     1.085
## 
## Concordance= 0.776  (se = 0.01 )
## Likelihood ratio test= 1066  on 16 df,   p=<2e-16
## Wald test            = 673.8  on 16 df,   p=<2e-16
## Score (logrank) test = 593.2  on 16 df,   p=<2e-16

91.3.4 IPF-Only NO3 - Combined Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ NO3 + dx_yr + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3 + dx_yr + 
##     site, data = All_IPF, id = ID, cluster = cohort)
## 
##   n= 123367, number of events= 2749 
##    (723 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se      z Pr(>|z|)    
## NO3     -0.055082  0.946407  0.071464  0.032913 -1.674 0.094211 .  
## dx_yr    0.102461  1.107894  0.007228  0.075557  1.356 0.175076    
## site02R  0.716687  2.047638  0.340074  0.075150  9.537  < 2e-16 ***
## site03R -0.257667  0.772852  0.347669  0.033202 -7.761 8.45e-15 ***
## site04R -0.149383  0.861239  0.460121  0.095850 -1.559 0.119113    
## site05R -0.282614  0.753811  0.373167  0.044810 -6.307 2.85e-10 ***
## site06R -0.030122  0.970327  0.325474  0.073079 -0.412 0.680201    
## site07R -0.265497  0.766825  0.313618  0.108258 -2.452 0.014189 *  
## site09R -0.011993  0.988079  0.341013  0.030999 -0.387 0.698845    
## site1    0.274407  1.315750  0.283710  0.107072  2.563 0.010382 *  
## site101 -0.077716  0.925227  0.296019  0.141757 -0.548 0.583532    
## site102 -0.226097  0.797641  0.298189  0.114292 -1.978 0.047903 *  
## site103  0.094436  1.099039  0.286317  0.088962  1.062 0.288450    
## site104 -0.190098  0.826878  0.294700  0.091467 -2.078 0.037680 *  
## site105  0.104458  1.110109  0.290712  0.149457  0.699 0.484604    
## site106 -0.167587  0.845703  0.291194  0.100077 -1.675 0.094015 .  
## site107  0.234657  1.264475  0.327753  0.160000  1.467 0.142483    
## site108  0.169447  1.184650  0.370441  0.207307  0.817 0.413715    
## site10R -0.154385  0.856942  0.385246  0.071695 -2.153 0.031290 *  
## site11R -0.058308  0.943360  0.299375  0.064010 -0.911 0.362340    
## site12R -0.056247  0.945305  0.338793  0.043534 -1.292 0.196347    
## site13R -0.270225  0.763208  0.303866  0.100445 -2.690 0.007139 ** 
## site14R -0.317286  0.728123  0.643776  0.123325 -2.573 0.010089 *  
## site15R  0.031964  1.032481  0.330466  0.014503  2.204 0.027526 *  
## site16R  0.174272  1.190379  0.347089  0.039028  4.465 8.00e-06 ***
## site17R  0.092022  1.096389  0.365664  0.029937  3.074 0.002113 ** 
## site18R -0.257438  0.773030  0.321665  0.159066 -1.618 0.105570    
## site19R -0.042549  0.958343  0.350157  0.009774 -4.353 1.34e-05 ***
## site20R -0.232595  0.792474  0.350990  0.042607 -5.459 4.79e-08 ***
## site21R -0.304105  0.737783  0.319500  0.051740 -5.878 4.16e-09 ***
## site22R -0.023130  0.977136  0.311019  0.037807 -0.612 0.540683    
## site23R -0.147031  0.863267  0.323173  0.024710 -5.950 2.68e-09 ***
## site24R -0.110673  0.895231  0.338152  0.066507 -1.664 0.096096 .  
## site25R  0.026198  1.026544  0.330693  0.058690  0.446 0.655320    
## site26R -0.410581  0.663265  0.344492  0.083157 -4.937 7.92e-07 ***
## site27R  0.027991  1.028387  0.761761  0.117887  0.237 0.812314    
## site28R -0.088808  0.915021  0.438947  0.050031 -1.775 0.075888 .  
## site29R -0.283223  0.753352  0.434674  0.053451 -5.299 1.17e-07 ***
## site30R -0.132965  0.875496  0.345753  0.112220 -1.185 0.236075    
## site31R -0.330245  0.718747  0.449491  0.071997 -4.587 4.50e-06 ***
## site32R -0.345744  0.707693  0.368594  0.037917 -9.118  < 2e-16 ***
## site33R -0.344040  0.708901  0.332316  0.089733 -3.834 0.000126 ***
## site34R -0.135915  0.872917  0.315011  0.050318 -2.701 0.006911 ** 
## site35R -0.110211  0.895645  0.333871  0.041637 -2.647 0.008122 ** 
## site36R -0.161577  0.850801  0.345970  0.053016 -3.048 0.002306 ** 
## site37R -0.128710  0.879229  0.341341  0.064060 -2.009 0.044514 *  
## site38R -0.292125  0.746675  0.328997  0.037171 -7.859 3.88e-15 ***
## site39R -0.185581  0.830622  0.362570  0.034017 -5.456 4.88e-08 ***
## site40R  0.280770  1.324149  0.370757  0.116725  2.405 0.016155 *  
## site41R -0.174710  0.839700  0.371677  0.048949 -3.569 0.000358 ***
## site42R  0.121714  1.129431  0.363395  0.033424  3.642 0.000271 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## NO3        0.9464     1.0566    0.8873    1.0095
## dx_yr      1.1079     0.9026    0.9554    1.2847
## site02R    2.0476     0.4884    1.7672    2.3726
## site03R    0.7729     1.2939    0.7242    0.8248
## site04R    0.8612     1.1611    0.7137    1.0392
## site05R    0.7538     1.3266    0.6904    0.8230
## site06R    0.9703     1.0306    0.8408    1.1198
## site07R    0.7668     1.3041    0.6202    0.9481
## site09R    0.9881     1.0121    0.9298    1.0500
## site1      1.3158     0.7600    1.0667    1.6230
## site101    0.9252     1.0808    0.7008    1.2216
## site102    0.7976     1.2537    0.6376    0.9979
## site103    1.0990     0.9099    0.9232    1.3084
## site104    0.8269     1.2094    0.6912    0.9892
## site105    1.1101     0.9008    0.8282    1.4879
## site106    0.8457     1.1824    0.6951    1.0290
## site107    1.2645     0.7908    0.9241    1.7302
## site108    1.1846     0.8441    0.7891    1.7785
## site10R    0.8569     1.1669    0.7446    0.9862
## site11R    0.9434     1.0600    0.8321    1.0695
## site12R    0.9453     1.0579    0.8680    1.0295
## site13R    0.7632     1.3103    0.6268    0.9293
## site14R    0.7281     1.3734    0.5718    0.9272
## site15R    1.0325     0.9685    1.0035    1.0623
## site16R    1.1904     0.8401    1.1027    1.2850
## site17R    1.0964     0.9121    1.0339    1.1626
## site18R    0.7730     1.2936    0.5660    1.0558
## site19R    0.9583     1.0435    0.9402    0.9769
## site20R    0.7925     1.2619    0.7290    0.8615
## site21R    0.7378     1.3554    0.6666    0.8165
## site22R    0.9771     1.0234    0.9073    1.0523
## site23R    0.8633     1.1584    0.8225    0.9061
## site24R    0.8952     1.1170    0.7858    1.0199
## site25R    1.0265     0.9741    0.9150    1.1517
## site26R    0.6633     1.5077    0.5635    0.7807
## site27R    1.0284     0.9724    0.8162    1.2957
## site28R    0.9150     1.0929    0.8296    1.0093
## site29R    0.7534     1.3274    0.6784    0.8366
## site30R    0.8755     1.1422    0.7026    1.0909
## site31R    0.7187     1.3913    0.6242    0.8277
## site32R    0.7077     1.4130    0.6570    0.7623
## site33R    0.7089     1.4106    0.5946    0.8452
## site34R    0.8729     1.1456    0.7909    0.9634
## site35R    0.8956     1.1165    0.8255    0.9718
## site36R    0.8508     1.1754    0.7668    0.9440
## site37R    0.8792     1.1374    0.7755    0.9968
## site38R    0.7467     1.3393    0.6942    0.8031
## site39R    0.8306     1.2039    0.7770    0.8879
## site40R    1.3241     0.7552    1.0534    1.6645
## site41R    0.8397     1.1909    0.7629    0.9243
## site42R    1.1294     0.8854    1.0578    1.2059
## 
## Concordance= 0.595  (se = 0.044 )
## Likelihood ratio test= 348.4  on 51 df,   p=<2e-16
## Wald test            = 3.92  on 51 df,   p=1
## Score (logrank) test = 337.7  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NO3 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NO3 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All_IPF, 
##     id = ID, cluster = cohort)
## 
##   n= 121241, number of events= 2709 
##    (2849 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## NO3                -0.054564  0.946898  0.072494  0.030757  -1.774 0.076057 .  
## dx_yr               0.099218  1.104307  0.007606  0.081135   1.223 0.221376    
## age_dx              0.004243  1.004252  0.002446  0.002134   1.988 0.046801 *  
## sexF               -0.159836  0.852283  0.044922  0.085770  -1.864 0.062386 .  
## dich_RaceNon-White  0.018578  1.018752  0.068696  0.061005   0.305 0.760723    
## smokeHxFormer       0.098557  1.103577  0.059375  0.039863   2.472 0.013422 *  
## smokeHxAlways      -0.030814  0.969656  0.138343  0.115025  -0.268 0.788788    
## smokeHxUnknown      0.321092  1.378633  0.153737  0.235711   1.362 0.173125    
## smokeHxEver         0.036511  1.037186  0.064244  0.008971   4.070 4.70e-05 ***
## disadv              0.042646  1.043569  0.070464  0.126598   0.337 0.736221    
## site02R             0.705684  2.025231  0.350996  0.102743   6.868 6.49e-12 ***
## site03R            -0.234234  0.791177  0.357870  0.040562  -5.775 7.71e-09 ***
## site04R            -0.142465  0.867218  0.468683  0.122649  -1.162 0.245414    
## site05R            -0.277113  0.757968  0.383738  0.032977  -8.403  < 2e-16 ***
## site06R            -0.040991  0.959838  0.338136  0.091358  -0.449 0.653661    
## site07R            -0.272326  0.761606  0.325345  0.118280  -2.302 0.021314 *  
## site09R            -0.058391  0.943281  0.353326  0.026494  -2.204 0.027527 *  
## site1               0.208889  1.232308  0.300071  0.135454   1.542 0.123041    
## site101            -0.127240  0.880522  0.311236  0.136469  -0.932 0.351142    
## site102            -0.294134  0.745177  0.313901  0.113372  -2.594 0.009475 ** 
## site103             0.019954  1.020154  0.301973  0.055022   0.363 0.716867    
## site104            -0.253621  0.775985  0.310452  0.063839  -3.973 7.10e-05 ***
## site105             0.047810  1.048971  0.307847  0.139797   0.342 0.732355    
## site106            -0.244867  0.782809  0.307248  0.090617  -2.702 0.006888 ** 
## site107             0.143008  1.153739  0.343367  0.143180   0.999 0.317892    
## site108             0.096605  1.101425  0.383039  0.193061   0.500 0.616804    
## site10R            -0.218385  0.803816  0.396007  0.051323  -4.255 2.09e-05 ***
## site11R            -0.090080  0.913858  0.313636  0.057871  -1.557 0.119574    
## site12R            -0.075753  0.927046  0.350124  0.034140  -2.219 0.026496 *  
## site13R            -0.306842  0.735767  0.315441  0.110162  -2.785 0.005346 ** 
## site14R            -0.321815  0.724832  0.649699  0.074691  -4.309 1.64e-05 ***
## site15R            -0.006817  0.993207  0.342528  0.006763  -1.008 0.313501    
## site16R             0.178820  1.195806  0.357268  0.025752   6.944 3.81e-12 ***
## site17R             0.079280  1.082508  0.376556  0.022848   3.470 0.000521 ***
## site18R            -0.286966  0.750537  0.333607  0.138826  -2.067 0.038726 *  
## site19R            -0.082713  0.920615  0.360774  0.011581  -7.142 9.19e-13 ***
## site20R            -0.227829  0.796260  0.362425  0.033471  -6.807 9.99e-12 ***
## site21R            -0.340566  0.711367  0.330833  0.046167  -7.377 1.62e-13 ***
## site22R            -0.032772  0.967759  0.323045  0.036951  -0.887 0.375136    
## site23R            -0.203361  0.815984  0.336895  0.019874 -10.233  < 2e-16 ***
## site24R            -0.121742  0.885376  0.349939  0.046920  -2.595 0.009467 ** 
## site25R             0.011184  1.011247  0.342676  0.082883   0.135 0.892663    
## site26R            -0.448746  0.638428  0.355354  0.062591  -7.170 7.53e-13 ***
## site27R             0.045162  1.046197  0.769068  0.135030   0.334 0.738032    
## site28R            -0.087361  0.916346  0.447924  0.039098  -2.234 0.025455 *  
## site29R            -0.286051  0.751224  0.443920  0.089187  -3.207 0.001340 ** 
## site30R            -0.194855  0.822954  0.358505  0.125477  -1.553 0.120443    
## site31R            -0.364271  0.694703  0.457648  0.068311  -5.333 9.69e-08 ***
## site32R            -0.383142  0.681716  0.388935  0.038088 -10.059  < 2e-16 ***
## site33R            -0.383915  0.681189  0.345400  0.069945  -5.489 4.05e-08 ***
## site34R            -0.149150  0.861440  0.327762  0.056229  -2.653 0.007989 ** 
## site35R            -0.159639  0.852452  0.346919  0.034180  -4.670 3.00e-06 ***
## site36R            -0.165895  0.847135  0.357402  0.046568  -3.562 0.000367 ***
## site37R            -0.143896  0.865978  0.352327  0.076814  -1.873 0.061028 .  
## site38R            -0.297396  0.742750  0.340216  0.022537 -13.196  < 2e-16 ***
## site39R            -0.225284  0.798290  0.373279  0.033474  -6.730 1.69e-11 ***
## site40R             0.238361  1.269167  0.381394  0.118740   2.007 0.044706 *  
## site41R            -0.212814  0.808306  0.381070  0.051994  -4.093 4.26e-05 ***
## site42R             0.098230  1.103217  0.373549  0.041853   2.347 0.018924 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NO3                   0.9469     1.0561    0.8915    1.0057
## dx_yr                 1.1043     0.9055    0.9419    1.2946
## age_dx                1.0043     0.9958    1.0001    1.0085
## sexF                  0.8523     1.1733    0.7204    1.0083
## dich_RaceNon-White    1.0188     0.9816    0.9039    1.1481
## smokeHxFormer         1.1036     0.9061    1.0206    1.1933
## smokeHxAlways         0.9697     1.0313    0.7739    1.2149
## smokeHxUnknown        1.3786     0.7254    0.8686    2.1882
## smokeHxEver           1.0372     0.9641    1.0191    1.0556
## disadv                1.0436     0.9583    0.8143    1.3375
## site02R               2.0252     0.4938    1.6558    2.4770
## site03R               0.7912     1.2639    0.7307    0.8566
## site04R               0.8672     1.1531    0.6819    1.1029
## site05R               0.7580     1.3193    0.7105    0.8086
## site06R               0.9598     1.0418    0.8025    1.1481
## site07R               0.7616     1.3130    0.6040    0.9603
## site09R               0.9433     1.0601    0.8955    0.9936
## site1                 1.2323     0.8115    0.9450    1.6070
## site101               0.8805     1.1357    0.6739    1.1505
## site102               0.7452     1.3420    0.5967    0.9306
## site103               1.0202     0.9802    0.9159    1.1363
## site104               0.7760     1.2887    0.6847    0.8794
## site105               1.0490     0.9533    0.7976    1.3796
## site106               0.7828     1.2775    0.6554    0.9350
## site107               1.1537     0.8667    0.8714    1.5275
## site108               1.1014     0.9079    0.7544    1.6080
## site10R               0.8038     1.2441    0.7269    0.8889
## site11R               0.9139     1.0943    0.8159    1.0236
## site12R               0.9270     1.0787    0.8670    0.9912
## site13R               0.7358     1.3591    0.5929    0.9131
## site14R               0.7248     1.3796    0.6261    0.8391
## site15R               0.9932     1.0068    0.9801    1.0065
## site16R               1.1958     0.8363    1.1369    1.2577
## site17R               1.0825     0.9238    1.0351    1.1321
## site18R               0.7505     1.3324    0.5717    0.9852
## site19R               0.9206     1.0862    0.9000    0.9418
## site20R               0.7963     1.2559    0.7457    0.8502
## site21R               0.7114     1.4057    0.6498    0.7787
## site22R               0.9678     1.0333    0.9001    1.0404
## site23R               0.8160     1.2255    0.7848    0.8484
## site24R               0.8854     1.1295    0.8076    0.9707
## site25R               1.0112     0.9889    0.8596    1.1896
## site26R               0.6384     1.5663    0.5647    0.7218
## site27R               1.0462     0.9558    0.8029    1.3632
## site28R               0.9163     1.0913    0.8487    0.9893
## site29R               0.7512     1.3312    0.6307    0.8947
## site30R               0.8230     1.2151    0.6435    1.0524
## site31R               0.6947     1.4395    0.6076    0.7942
## site32R               0.6817     1.4669    0.6327    0.7346
## site33R               0.6812     1.4680    0.5939    0.7813
## site34R               0.8614     1.1608    0.7715    0.9618
## site35R               0.8525     1.1731    0.7972    0.9115
## site36R               0.8471     1.1804    0.7732    0.9281
## site37R               0.8660     1.1548    0.7449    1.0067
## site38R               0.7427     1.3463    0.7107    0.7763
## site39R               0.7983     1.2527    0.7476    0.8524
## site40R               1.2692     0.7879    1.0057    1.6017
## site41R               0.8083     1.2372    0.7300    0.8950
## site42R               1.1032     0.9064    1.0163    1.1975
## 
## Concordance= 0.598  (se = 0.039 )
## Likelihood ratio test= 380.4  on 59 df,   p=<2e-16
## Wald test            = 5.61  on 59 df,   p=1
## Score (logrank) test = 374.6  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

91.4 IPF-Only NH4

91.4.1 IPF-Only NH4 - Simmons Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ NH4 + dx_yr, data=Simm_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4 + dx_yr, data = Simm_IPF, 
##     id = ID)
## 
##   n= 31982, number of events= 695 
##    (446 observations deleted due to missingness)
## 
##          coef exp(coef) se(coef)     z Pr(>|z|)  
## NH4   0.15537   1.16809  0.13567 1.145   0.2521  
## dx_yr 0.03761   1.03832  0.01508 2.494   0.0126 *
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##       exp(coef) exp(-coef) lower .95 upper .95
## NH4       1.168     0.8561    0.8953     1.524
## dx_yr     1.038     0.9631    1.0081     1.069
## 
## Concordance= 0.522  (se = 0.013 )
## Likelihood ratio test= 8.91  on 2 df,   p=0.01
## Wald test            = 8.76  on 2 df,   p=0.01
## Score (logrank) test = 8.78  on 2 df,   p=0.01

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NH4 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv, data=Simm_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv, data = Simm_IPF, 
##     id = ID)
## 
##   n= 30488, number of events= 670 
##    (1940 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## NH4                 0.157834  1.170972  0.142271  1.109  0.26726    
## dx_yr               0.025811  1.026147  0.016055  1.608  0.10791    
## age_dx              0.001899  1.001900  0.004559  0.416  0.67709    
## sexF               -0.377828  0.685349  0.088779 -4.256 2.08e-05 ***
## dich_RaceNon-White  0.164438  1.178730  0.128565  1.279  0.20089    
## smokeHxFormer       0.082867  1.086398  0.091966  0.901  0.36755    
## smokeHxAlways      -0.403611  0.667904  0.266618 -1.514  0.13007    
## smokeHxUnknown      0.496943  1.643689  0.169764  2.927  0.00342 ** 
## disadv              0.382206  1.465514  0.134148  2.849  0.00438 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NH4                   1.1710     0.8540    0.8860    1.5476
## dx_yr                 1.0261     0.9745    0.9944    1.0589
## age_dx                1.0019     0.9981    0.9930    1.0109
## sexF                  0.6853     1.4591    0.5759    0.8156
## dich_RaceNon-White    1.1787     0.8484    0.9162    1.5165
## smokeHxFormer         1.0864     0.9205    0.9072    1.3010
## smokeHxAlways         0.6679     1.4972    0.3961    1.1263
## smokeHxUnknown        1.6437     0.6084    1.1785    2.2926
## disadv                1.4655     0.6824    1.1267    1.9062
## 
## Concordance= 0.591  (se = 0.012 )
## Likelihood ratio test= 52.86  on 9 df,   p=3e-08
## Wald test            = 53.21  on 9 df,   p=3e-08
## Score (logrank) test = 53.95  on 9 df,   p=2e-08

91.4.2 IPF-Only NH4 - PFF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ NH4 + dx_yr + site, data=PFF_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4 + dx_yr + 
##     site, data = PFF_IPF, id = ID)
## 
##   n= 54499, number of events= 1146 
##    (217 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef)      z Pr(>|z|)    
## NH4      0.299578  1.349289  0.205870  1.455   0.1456    
## dx_yr    0.128522  1.137147  0.021647  5.937  2.9e-09 ***
## site02R  0.563769  1.757282  0.341884  1.649   0.0991 .  
## site03R -0.267417  0.765354  0.347439 -0.770   0.4415    
## site04R -0.389703  0.677258  0.466576 -0.835   0.4036    
## site05R -0.506124  0.602827  0.395179 -1.281   0.2003    
## site06R -0.133983  0.874605  0.334200 -0.401   0.6885    
## site07R -0.475317  0.621688  0.327158 -1.453   0.1463    
## site08R        NA        NA  0.000000     NA       NA    
## site09R -0.116651  0.889895  0.347165 -0.336   0.7369    
## site10R -0.141840  0.867760  0.385315 -0.368   0.7128    
## site11R -0.126524  0.881153  0.303209 -0.417   0.6765    
## site12R -0.179617  0.835590  0.340895 -0.527   0.5983    
## site13R -0.420066  0.657004  0.315697 -1.331   0.1833    
## site14R -0.525120  0.591484  0.652916 -0.804   0.4212    
## site15R -0.001739  0.998262  0.331854 -0.005   0.9958    
## site16R  0.024294  1.024591  0.353837  0.069   0.9453    
## site17R -0.120791  0.886219  0.377038 -0.320   0.7487    
## site18R -0.340826  0.711182  0.326550 -1.044   0.2966    
## site19R  0.009459  1.009504  0.350178  0.027   0.9785    
## site20R -0.358113  0.698994  0.355973 -1.006   0.3144    
## site21R -0.358986  0.698384  0.324012 -1.108   0.2679    
## site22R -0.220351  0.802237  0.326009 -0.676   0.4991    
## site23R -0.244747  0.782902  0.328677 -0.745   0.4565    
## site24R -0.251247  0.777830  0.340470 -0.738   0.4605    
## site25R -0.091436  0.912620  0.330747 -0.276   0.7822    
## site26R -0.484024  0.616299  0.350134 -1.382   0.1669    
## site27R -0.152646  0.858434  0.769209 -0.198   0.8427    
## site28R -0.322003  0.724696  0.457481 -0.704   0.4815    
## site29R -0.374218  0.687827  0.439059 -0.852   0.3940    
## site30R -0.395214  0.673536  0.357303 -1.106   0.2687    
## site31R -0.397047  0.672302  0.453386 -0.876   0.3812    
## site32R -0.354642  0.701425  0.368645 -0.962   0.3360    
## site33R -0.428317  0.651605  0.339143 -1.263   0.2066    
## site34R -0.255439  0.774577  0.323685 -0.789   0.4300    
## site35R -0.160720  0.851530  0.336006 -0.478   0.6324    
## site36R -0.410755  0.663149  0.370201 -1.110   0.2672    
## site37R -0.367693  0.692329  0.363746 -1.011   0.3121    
## site38R -0.352701  0.702787  0.332979 -1.059   0.2895    
## site39R -0.252164  0.777118  0.361934 -0.697   0.4860    
## site40R -0.061528  0.940327  0.379882 -0.162   0.8713    
## site41R -0.320767  0.725592  0.378005 -0.849   0.3961    
## site42R -0.073911  0.928754  0.376203 -0.196   0.8442    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## NH4        1.3493     0.7411    0.9013     2.020
## dx_yr      1.1371     0.8794    1.0899     1.186
## site02R    1.7573     0.5691    0.8991     3.434
## site03R    0.7654     1.3066    0.3874     1.512
## site04R    0.6773     1.4765    0.2714     1.690
## site05R    0.6028     1.6588    0.2779     1.308
## site06R    0.8746     1.1434    0.4543     1.684
## site07R    0.6217     1.6085    0.3274     1.180
## site08R        NA         NA        NA        NA
## site09R    0.8899     1.1237    0.4506     1.757
## site10R    0.8678     1.1524    0.4078     1.847
## site11R    0.8812     1.1349    0.4864     1.596
## site12R    0.8356     1.1968    0.4284     1.630
## site13R    0.6570     1.5221    0.3539     1.220
## site14R    0.5915     1.6907    0.1645     2.127
## site15R    0.9983     1.0017    0.5209     1.913
## site16R    1.0246     0.9760    0.5121     2.050
## site17R    0.8862     1.1284    0.4233     1.856
## site18R    0.7112     1.4061    0.3750     1.349
## site19R    1.0095     0.9906    0.5082     2.005
## site20R    0.6990     1.4306    0.3479     1.404
## site21R    0.6984     1.4319    0.3701     1.318
## site22R    0.8022     1.2465    0.4235     1.520
## site23R    0.7829     1.2773    0.4111     1.491
## site24R    0.7778     1.2856    0.3991     1.516
## site25R    0.9126     1.0957    0.4773     1.745
## site26R    0.6163     1.6226    0.3103     1.224
## site27R    0.8584     1.1649    0.1901     3.877
## site28R    0.7247     1.3799    0.2956     1.776
## site29R    0.6878     1.4539    0.2909     1.626
## site30R    0.6735     1.4847    0.3344     1.357
## site31R    0.6723     1.4874    0.2765     1.635
## site32R    0.7014     1.4257    0.3406     1.445
## site33R    0.6516     1.5347    0.3352     1.267
## site34R    0.7746     1.2910    0.4107     1.461
## site35R    0.8515     1.1744    0.4407     1.645
## site36R    0.6631     1.5080    0.3210     1.370
## site37R    0.6923     1.4444    0.3394     1.412
## site38R    0.7028     1.4229    0.3659     1.350
## site39R    0.7771     1.2868    0.3823     1.580
## site40R    0.9403     1.0635    0.4466     1.980
## site41R    0.7256     1.3782    0.3459     1.522
## site42R    0.9288     1.0767    0.4443     1.941
## 
## Concordance= 0.59  (se = 0.009 )
## Likelihood ratio test= 112.6  on 42 df,   p=2e-08
## Wald test            = 111.5  on 42 df,   p=3e-08
## Score (logrank) test = 115.5  on 42 df,   p=9e-09

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NH4 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=PFF_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = PFF_IPF, 
##     id = ID)
## 
##   n= 53867, number of events= 1131 
##    (849 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)      z Pr(>|z|)    
## NH4                 0.3099396  1.3633427  0.2089333  1.483    0.138    
## dx_yr               0.1282427  1.1368289  0.0222315  5.769    8e-09 ***
## age_dx             -0.0002757  0.9997243  0.0039383 -0.070    0.944    
## sexM                0.0603721  1.0622317  0.0713335  0.846    0.397    
## dich_RaceNon-White -0.0675310  0.9346988  0.1263415 -0.535    0.593    
## smokeHxEver         0.0425310  1.0434484  0.0646444  0.658    0.511    
## disadv             -0.0451066  0.9558956  0.1121488 -0.402    0.688    
## site02R             0.5110305  1.6670081  0.3544584  1.442    0.149    
## site03R            -0.2976238  0.7425807  0.3597288 -0.827    0.408    
## site04R            -0.4486624  0.6384816  0.4774194 -0.940    0.347    
## site05R            -0.5561066  0.5734374  0.4078989 -1.363    0.173    
## site06R            -0.1744525  0.8399168  0.3501181 -0.498    0.618    
## site07R            -0.5229806  0.5927512  0.3414992 -1.531    0.126    
## site08R                    NA         NA  0.0000000     NA       NA    
## site09R            -0.1988847  0.8196444  0.3617123 -0.550    0.582    
## site10R            -0.1998395  0.8188622  0.3985339 -0.501    0.616    
## site11R            -0.1880904  0.8285398  0.3213644 -0.585    0.558    
## site12R            -0.2278811  0.7962189  0.3551676 -0.642    0.521    
## site13R            -0.4801375  0.6186983  0.3290952 -1.459    0.145    
## site14R            -0.5908412  0.5538612  0.6606571 -0.894    0.371    
## site15R            -0.0526096  0.9487504  0.3474469 -0.151    0.880    
## site16R            -0.0217557  0.9784793  0.3651909 -0.060    0.952    
## site17R            -0.1826296  0.8330767  0.3907721 -0.467    0.640    
## site18R            -0.3788716  0.6846335  0.3421525 -1.107    0.268    
## site19R            -0.0369094  0.9637634  0.3625156 -0.102    0.919    
## site20R            -0.4132719  0.6614824  0.3700763 -1.117    0.264    
## site21R            -0.4042978  0.6674453  0.3376549 -1.197    0.231    
## site22R            -0.2773252  0.7578080  0.3402512 -0.815    0.415    
## site23R            -0.3119271  0.7320349  0.3445775 -0.905    0.365    
## site24R            -0.3101700  0.7333223  0.3544824 -0.875    0.382    
## site25R            -0.1289293  0.8790361  0.3442595 -0.375    0.708    
## site26R            -0.5212456  0.5937805  0.3634313 -1.434    0.152    
## site27R            -0.2609816  0.7702951  0.7804243 -0.334    0.738    
## site28R            -0.3607865  0.6971278  0.4697816 -0.768    0.442    
## site29R            -0.4300704  0.6504633  0.4505923 -0.954    0.340    
## site30R            -0.4558110  0.6339337  0.3706741 -1.230    0.219    
## site31R            -0.4493241  0.6380593  0.4625708 -0.971    0.331    
## site32R            -0.4218615  0.6558249  0.3905394 -1.080    0.280    
## site33R            -0.4697900  0.6251335  0.3545632 -1.325    0.185    
## site34R            -0.3022780  0.7391326  0.3393083 -0.891    0.373    
## site35R            -0.2220664  0.8008622  0.3504066 -0.634    0.526    
## site36R            -0.4512105  0.6368568  0.3854093 -1.171    0.242    
## site37R            -0.4167212  0.6592047  0.3768232 -1.106    0.269    
## site38R            -0.3925347  0.6753429  0.3466539 -1.132    0.257    
## site39R            -0.3055217  0.7367389  0.3745333 -0.816    0.415    
## site40R            -0.0973459  0.9072421  0.3906020 -0.249    0.803    
## site41R            -0.3619630  0.6963081  0.3885240 -0.932    0.352    
## site42R            -0.1427495  0.8669712  0.3884525 -0.367    0.713    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NH4                   1.3633     0.7335    0.9052     2.053
## dx_yr                 1.1368     0.8796    1.0884     1.187
## age_dx                0.9997     1.0003    0.9920     1.007
## sexM                  1.0622     0.9414    0.9236     1.222
## dich_RaceNon-White    0.9347     1.0699    0.7297     1.197
## smokeHxEver           1.0434     0.9584    0.9193     1.184
## disadv                0.9559     1.0461    0.7673     1.191
## site02R               1.6670     0.5999    0.8322     3.339
## site03R               0.7426     1.3467    0.3669     1.503
## site04R               0.6385     1.5662    0.2505     1.628
## site05R               0.5734     1.7439    0.2578     1.276
## site06R               0.8399     1.1906    0.4229     1.668
## site07R               0.5928     1.6870    0.3035     1.158
## site08R                   NA         NA        NA        NA
## site09R               0.8196     1.2200    0.4034     1.665
## site10R               0.8189     1.2212    0.3750     1.788
## site11R               0.8285     1.2069    0.4413     1.555
## site12R               0.7962     1.2559    0.3969     1.597
## site13R               0.6187     1.6163    0.3246     1.179
## site14R               0.5539     1.8055    0.1517     2.022
## site15R               0.9488     1.0540    0.4802     1.875
## site16R               0.9785     1.0220    0.4783     2.002
## site17R               0.8331     1.2004    0.3873     1.792
## site18R               0.6846     1.4606    0.3501     1.339
## site19R               0.9638     1.0376    0.4736     1.961
## site20R               0.6615     1.5118    0.3203     1.366
## site21R               0.6674     1.4983    0.3444     1.294
## site22R               0.7578     1.3196    0.3890     1.476
## site23R               0.7320     1.3661    0.3726     1.438
## site24R               0.7333     1.3637    0.3661     1.469
## site25R               0.8790     1.1376    0.4477     1.726
## site26R               0.5938     1.6841    0.2913     1.211
## site27R               0.7703     1.2982    0.1669     3.556
## site28R               0.6971     1.4345    0.2776     1.751
## site29R               0.6505     1.5374    0.2690     1.573
## site30R               0.6339     1.5775    0.3066     1.311
## site31R               0.6381     1.5673    0.2577     1.580
## site32R               0.6558     1.5248    0.3050     1.410
## site33R               0.6251     1.5997    0.3120     1.252
## site34R               0.7391     1.3529    0.3801     1.437
## site35R               0.8009     1.2487    0.4030     1.592
## site36R               0.6369     1.5702    0.2992     1.356
## site37R               0.6592     1.5170    0.3150     1.380
## site38R               0.6753     1.4807    0.3423     1.332
## site39R               0.7367     1.3573    0.3536     1.535
## site40R               0.9072     1.1022    0.4219     1.951
## site41R               0.6963     1.4361    0.3252     1.491
## site42R               0.8670     1.1534    0.4049     1.856
## 
## Concordance= 0.591  (se = 0.009 )
## Likelihood ratio test= 113.4  on 47 df,   p=2e-07
## Wald test            = 112.6  on 47 df,   p=3e-07
## Score (logrank) test = 116.6  on 47 df,   p=8e-08

91.4.3 IPF-Only NH4 - CARE-PF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ NH4 + dx_yr + site, data=CARE_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4 + dx_yr + 
##     site, data = CARE_IPF, id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef)      z Pr(>|z|)    
## NH4     -2.32104   0.09817  0.43145 -5.380 7.46e-08 ***
## dx_yr    0.72763   2.07017  0.03556 20.463  < 2e-16 ***
## site102 -0.35162   0.70355  0.17863 -1.968  0.04902 *  
## site103 -0.19000   0.82696  0.17429 -1.090  0.27567    
## site104 -0.33236   0.71723  0.18811 -1.767  0.07726 .  
## site105 -0.12298   0.88429  0.12980 -0.947  0.34341    
## site106  0.35328   1.42373  0.12770  2.767  0.00566 ** 
## site107 -0.04042   0.96038  0.19850 -0.204  0.83863    
## site108 -0.81867   0.44102  0.27338 -2.995  0.00275 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## NH4       0.09817    10.1862   0.04214    0.2287
## dx_yr     2.07017     0.4831   1.93080    2.2196
## site102   0.70355     1.4214   0.49573    0.9985
## site103   0.82696     1.2092   0.58766    1.1637
## site104   0.71723     1.3943   0.49606    1.0370
## site105   0.88429     1.1309   0.68566    1.1405
## site106   1.42373     0.7024   1.10849    1.8286
## site107   0.96038     1.0413   0.65085    1.4171
## site108   0.44102     2.2675   0.25808    0.7536
## 
## Concordance= 0.777  (se = 0.01 )
## Likelihood ratio test= 1080  on 9 df,   p=<2e-16
## Wald test            = 691.6  on 9 df,   p=<2e-16
## Score (logrank) test = 586.3  on 9 df,   p=<2e-16

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NH4 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=CARE_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = CARE_IPF, 
##     id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## NH4                -2.417776  0.089120  0.437706 -5.524 3.32e-08 ***
## dx_yr               0.723758  2.062168  0.035768 20.235  < 2e-16 ***
## age_dx              0.013063  1.013149  0.004556  2.867  0.00414 ** 
## sexF               -0.048076  0.953061  0.077150 -0.623  0.53318    
## dich_RaceNon-White -0.142324  0.867340  0.110735 -1.285  0.19870    
## smokeHxFormer      -0.016654  0.983483  0.081456 -0.204  0.83799    
## smokeHxAlways       0.054267  1.055767  0.168460  0.322  0.74735    
## smokeHxUnknown      0.435943  1.546421  0.734682  0.593  0.55293    
## disadv              0.022015  1.022259  0.130595  0.169  0.86613    
## site102            -0.393999  0.674355  0.181678 -2.169  0.03011 *  
## site103            -0.209966  0.810611  0.179209 -1.172  0.24134    
## site104            -0.356652  0.700016  0.190645 -1.871  0.06138 .  
## site105            -0.114527  0.891788  0.133049 -0.861  0.38936    
## site106             0.325792  1.385128  0.128596  2.533  0.01129 *  
## site107            -0.118235  0.888487  0.206228 -0.573  0.56643    
## site108            -0.839694  0.431843  0.278327 -3.017  0.00255 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NH4                  0.08912    11.2209   0.03779    0.2102
## dx_yr                2.06217     0.4849   1.92255    2.2119
## age_dx               1.01315     0.9870   1.00414    1.0222
## sexF                 0.95306     1.0493   0.81932    1.1086
## dich_RaceNon-White   0.86734     1.1530   0.69812    1.0776
## smokeHxFormer        0.98348     1.0168   0.83836    1.1537
## smokeHxAlways        1.05577     0.9472   0.75888    1.4688
## smokeHxUnknown       1.54642     0.6467   0.36641    6.5266
## disadv               1.02226     0.9782   0.79140    1.3205
## site102              0.67435     1.4829   0.47233    0.9628
## site103              0.81061     1.2336   0.57052    1.1517
## site104              0.70002     1.4285   0.48176    1.0172
## site105              0.89179     1.1213   0.68708    1.1575
## site106              1.38513     0.7220   1.07654    1.7822
## site107              0.88849     1.1255   0.59307    1.3310
## site108              0.43184     2.3157   0.25027    0.7451
## 
## Concordance= 0.778  (se = 0.01 )
## Likelihood ratio test= 1092  on 16 df,   p=<2e-16
## Wald test            = 698.9  on 16 df,   p=<2e-16
## Score (logrank) test = 593.3  on 16 df,   p=<2e-16

91.4.4 IPF-Only NH4 - Combined Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ NH4 + dx_yr + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4 + dx_yr + 
##     site, data = All_IPF, id = ID, cluster = cohort)
## 
##   n= 123367, number of events= 2749 
##    (723 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef) robust se      z Pr(>|z|)    
## NH4      0.93181   2.53910  0.10745   0.59258  1.572 0.115847    
## dx_yr    0.18588   1.20428  0.01202   0.12022  1.546 0.122068    
## site02R  0.36071   1.43435  0.33619   0.23780  1.517 0.129299    
## site03R -0.34668   0.70703  0.34717   0.07548 -4.593 4.37e-06 ***
## site04R -0.79110   0.45335  0.45390   0.39835 -1.986 0.047039 *  
## site05R -0.96392   0.38139  0.37555   0.36876 -2.614 0.008949 ** 
## site06R -0.38543   0.68016  0.32729   0.13282 -2.902 0.003708 ** 
## site07R -0.86156   0.42250  0.31083   0.40396 -2.133 0.032941 *  
## site09R -0.35739   0.69950  0.34172   0.18457 -1.936 0.052827 .  
## site1   -0.42537   0.65353  0.29210   0.30974 -1.373 0.169655    
## site101 -0.51725   0.59616  0.29687   0.35670 -1.450 0.147026    
## site102 -0.33333   0.71653  0.29810   0.14373 -2.319 0.020391 *  
## site103  0.07895   1.08215  0.28631   0.07879  1.002 0.316330    
## site104 -0.22962   0.79483  0.29475   0.10129 -2.267 0.023388 *  
## site105 -0.25463   0.77520  0.29203   0.31408 -0.811 0.417528    
## site106 -0.70545   0.49389  0.29172   0.37095 -1.902 0.057206 .  
## site107 -0.14332   0.86648  0.32915   0.33046 -0.434 0.664510    
## site108 -0.12451   0.88293  0.37065   0.32331 -0.385 0.700162    
## site10R -0.14188   0.86773  0.38522   0.05019 -2.827 0.004699 ** 
## site11R -0.27530   0.75935  0.30027   0.16124 -1.707 0.087750 .  
## site12R -0.40518   0.66686  0.33564   0.15484 -2.617 0.008876 ** 
## site13R -0.71628   0.48857  0.30540   0.31729 -2.257 0.023978 *  
## site14R -0.91750   0.39951  0.64423   0.21534 -4.261 2.04e-05 ***
## site15R -0.09940   0.90538  0.33074   0.07158 -1.389 0.164939    
## site16R -0.22061   0.80203  0.34675   0.16603 -1.329 0.183944    
## site17R -0.46256   0.62967  0.36445   0.28866 -1.602 0.109059    
## site18R -0.50355   0.60438  0.32270   0.26858 -1.875 0.060807 .  
## site19R  0.02740   1.02778  0.34998   0.04060  0.675 0.499785    
## site20R -0.62287   0.53640  0.34928   0.20077 -3.102 0.001920 ** 
## site21R -0.55516   0.57398  0.32048   0.16616 -3.341 0.000835 ***
## site22R -0.58779   0.55555  0.31087   0.29454 -1.996 0.045975 *  
## site23R -0.44893   0.63831  0.32397   0.16762 -2.678 0.007401 ** 
## site24R -0.51735   0.59610  0.33300   0.16627 -3.111 0.001862 ** 
## site25R -0.31624   0.72888  0.32700   0.24015 -1.317 0.187892    
## site26R -0.70005   0.49656  0.34604   0.25294 -2.768 0.005646 ** 
## site27R -0.56310   0.56944  0.76217   0.41860 -1.345 0.178561    
## site28R -0.79281   0.45257  0.44002   0.37497 -2.114 0.034487 *  
## site29R -0.61067   0.54299  0.43499   0.20375 -2.997 0.002725 ** 
## site30R -0.82606   0.43777  0.33778   0.43788 -1.887 0.059227 .  
## site31R -0.60412   0.54655  0.45039   0.20443 -2.955 0.003126 ** 
## site32R -0.39733   0.67211  0.36854   0.06318 -6.289 3.20e-10 ***
## site33R -0.66478   0.51438  0.33403   0.26183 -2.539 0.011117 *  
## site34R -0.52304   0.59272  0.31602   0.24352 -2.148 0.031729 *  
## site35R -0.28676   0.75069  0.33429   0.11336 -2.530 0.011419 *  
## site36R -0.89950   0.40677  0.34633   0.39776 -2.261 0.023733 *  
## site37R -0.83734   0.43286  0.34230   0.40559 -2.065 0.038967 *  
## site38R -0.52747   0.59010  0.32992   0.14344 -3.677 0.000236 ***
## site39R -0.38544   0.68015  0.36051   0.10887 -3.540 0.000400 ***
## site40R -0.58460   0.55733  0.35355   0.55890 -1.046 0.295572    
## site41R -0.59247   0.55296  0.37094   0.23678 -2.502 0.012342 *  
## site42R -0.40822   0.66483  0.36414   0.29518 -1.383 0.166681    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## NH4        2.5391     0.3938    0.7948    8.1113
## dx_yr      1.2043     0.8304    0.9515    1.5243
## site02R    1.4343     0.6972    0.9000    2.2860
## site03R    0.7070     1.4144    0.6098    0.8198
## site04R    0.4533     2.2058    0.2077    0.9897
## site05R    0.3814     2.6220    0.1851    0.7857
## site06R    0.6802     1.4703    0.5243    0.8824
## site07R    0.4225     2.3669    0.1914    0.9326
## site09R    0.6995     1.4296    0.4872    1.0044
## site1      0.6535     1.5302    0.3561    1.1993
## site101    0.5962     1.6774    0.2963    1.1994
## site102    0.7165     1.3956    0.5406    0.9497
## site103    1.0822     0.9241    0.9273    1.2629
## site104    0.7948     1.2581    0.6517    0.9694
## site105    0.7752     1.2900    0.4189    1.4347
## site106    0.4939     2.0248    0.2387    1.0218
## site107    0.8665     1.1541    0.4534    1.6560
## site108    0.8829     1.1326    0.4685    1.6639
## site10R    0.8677     1.1524    0.7864    0.9574
## site11R    0.7593     1.3169    0.5536    1.0416
## site12R    0.6669     1.4996    0.4923    0.9033
## site13R    0.4886     2.0468    0.2623    0.9099
## site14R    0.3995     2.5030    0.2620    0.6093
## site15R    0.9054     1.1045    0.7869    1.0417
## site16R    0.8020     1.2468    0.5792    1.1105
## site17R    0.6297     1.5881    0.3576    1.1087
## site18R    0.6044     1.6546    0.3570    1.0231
## site19R    1.0278     0.9730    0.9492    1.1129
## site20R    0.5364     1.8643    0.3619    0.7950
## site21R    0.5740     1.7422    0.4144    0.7949
## site22R    0.5556     1.8000    0.3119    0.9896
## site23R    0.6383     1.5666    0.4596    0.8866
## site24R    0.5961     1.6776    0.4303    0.8257
## site25R    0.7289     1.3720    0.4552    1.1670
## site26R    0.4966     2.0139    0.3025    0.8152
## site27R    0.5694     1.7561    0.2507    1.2935
## site28R    0.4526     2.2096    0.2170    0.9438
## site29R    0.5430     1.8417    0.3642    0.8095
## site30R    0.4378     2.2843    0.1856    1.0327
## site31R    0.5466     1.8296    0.3661    0.8159
## site32R    0.6721     1.4878    0.5938    0.7607
## site33R    0.5144     1.9441    0.3079    0.8593
## site34R    0.5927     1.6872    0.3678    0.9553
## site35R    0.7507     1.3321    0.6011    0.9375
## site36R    0.4068     2.4584    0.1865    0.8870
## site37R    0.4329     2.3102    0.1955    0.9585
## site38R    0.5901     1.6946    0.4455    0.7817
## site39R    0.6802     1.4703    0.5495    0.8419
## site40R    0.5573     1.7943    0.1864    1.6667
## site41R    0.5530     1.8084    0.3477    0.8795
## site42R    0.6648     1.5041    0.3728    1.1857
## 
## Concordance= 0.602  (se = 0.043 )
## Likelihood ratio test= 421.9  on 51 df,   p=<2e-16
## Wald test            = 2.47  on 51 df,   p=1
## Score (logrank) test = 394.6  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ NH4 + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ NH4 + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All_IPF, 
##     id = ID, cluster = cohort)
## 
##   n= 121241, number of events= 2709 
##    (2849 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se      z Pr(>|z|)    
## NH4                 0.967106  2.630322  0.109539  0.610140  1.585 0.112954    
## dx_yr               0.186356  1.204851  0.012442  0.126113  1.478 0.139491    
## age_dx              0.003553  1.003559  0.002438  0.001808  1.965 0.049390 *  
## sexF               -0.142803  0.866925  0.044902  0.085819 -1.664 0.096113 .  
## dich_RaceNon-White  0.015770  1.015895  0.068933  0.071316  0.221 0.824992    
## smokeHxFormer       0.094836  1.099479  0.059305  0.022491  4.217 2.48e-05 ***
## smokeHxAlways      -0.018509  0.981661  0.138598  0.099732 -0.186 0.852768    
## smokeHxUnknown      0.303964  1.355220  0.152997  0.195234  1.557 0.119489    
## smokeHxEver         0.034896  1.035512  0.064259  0.004294  8.127 4.41e-16 ***
## disadv              0.052803  1.054222  0.070526  0.133998  0.394 0.693537    
## site02R             0.338177  1.402389  0.346886  0.251039  1.347 0.177945    
## site03R            -0.325112  0.722446  0.357472  0.083079 -3.913 9.10e-05 ***
## site04R            -0.798887  0.449829  0.462200  0.417239 -1.915 0.055531 .  
## site05R            -0.981103  0.374898  0.385961  0.371306 -2.642 0.008234 ** 
## site06R            -0.406819  0.665765  0.340081  0.105357 -3.861 0.000113 ***
## site07R            -0.883017  0.413533  0.322387  0.404345 -2.184 0.028975 *  
## site09R            -0.409306  0.664111  0.353977  0.178265 -2.296 0.021673 *  
## site1              -0.514475  0.597814  0.308255  0.286044 -1.799 0.072084 .  
## site101            -0.579353  0.560261  0.311843  0.351149 -1.650 0.098968 .  
## site102            -0.396634  0.672580  0.313941  0.146336 -2.710 0.006720 ** 
## site103             0.008494  1.008530  0.301975  0.048669  0.175 0.861456    
## site104            -0.291553  0.747102  0.310558  0.078843 -3.698 0.000217 ***
## site105            -0.321905  0.724767  0.309137  0.303786 -1.060 0.289306    
## site106            -0.794692  0.451720  0.307490  0.362859 -2.190 0.028518 *  
## site107            -0.243152  0.784152  0.344712  0.315544 -0.771 0.440955    
## site108            -0.202172  0.816954  0.382962  0.310351 -0.651 0.514769    
## site10R            -0.198144  0.820252  0.395988  0.027688 -7.156 8.29e-13 ***
## site11R            -0.318822  0.727005  0.314582  0.151530 -2.104 0.035377 *  
## site12R            -0.432710  0.648749  0.346808  0.146472 -2.954 0.003135 ** 
## site13R            -0.761453  0.466987  0.316862  0.320768 -2.374 0.017604 *  
## site14R            -0.942307  0.389728  0.650155  0.230033 -4.096 4.20e-05 ***
## site15R            -0.142418  0.867259  0.342924  0.049604 -2.871 0.004091 ** 
## site16R            -0.229185  0.795181  0.356837  0.154980 -1.479 0.139191    
## site17R            -0.489918  0.612677  0.375235  0.287169 -1.706 0.088003 .  
## site18R            -0.540658  0.582365  0.334656  0.243692 -2.219 0.026513 *  
## site19R            -0.008249  0.991785  0.360654  0.052008 -0.159 0.873970    
## site20R            -0.630472  0.532340  0.360593  0.202499 -3.113 0.001849 ** 
## site21R            -0.595138  0.551486  0.331776  0.153498 -3.877 0.000106 ***
## site22R            -0.610871  0.542878  0.322627  0.301676 -2.025 0.042875 *  
## site23R            -0.512360  0.599080  0.337655  0.148170 -3.458 0.000544 ***
## site24R            -0.535427  0.585419  0.344579  0.156175 -3.428 0.000607 ***
## site25R            -0.350178  0.704563  0.338823  0.263175 -1.331 0.183324    
## site26R            -0.746592  0.473979  0.356864  0.231256 -3.228 0.001245 ** 
## site27R            -0.562870  0.569572  0.769697  0.426023 -1.321 0.186428    
## site28R            -0.812840  0.443596  0.448808  0.368002 -2.209 0.027189 *  
## site29R            -0.619752  0.538078  0.444140  0.236344 -2.622 0.008735 ** 
## site30R            -0.908980  0.402935  0.349699  0.438725 -2.072 0.038278 *  
## site31R            -0.642008  0.526235  0.458496  0.197200 -3.256 0.001131 ** 
## site32R            -0.458704  0.632103  0.388757  0.068145 -6.731 1.68e-11 ***
## site33R            -0.698631  0.497266  0.346948  0.230433 -3.032 0.002431 ** 
## site34R            -0.544555  0.580100  0.328612  0.246292 -2.211 0.027035 *  
## site35R            -0.336813  0.714042  0.347214  0.101148 -3.330 0.000869 ***
## site36R            -0.926335  0.396002  0.357491  0.394576 -2.348 0.018891 *  
## site37R            -0.868827  0.419443  0.352959  0.414290 -2.097 0.035981 *  
## site38R            -0.539272  0.583172  0.341158  0.121181 -4.450 8.58e-06 ***
## site39R            -0.428334  0.651594  0.371070  0.097419 -4.397 1.10e-05 ***
## site40R            -0.654111  0.519904  0.363493  0.549066 -1.191 0.233530    
## site41R            -0.638739  0.527958  0.380139  0.227405 -2.809 0.004972 ** 
## site42R            -0.442818  0.642224  0.374191  0.299865 -1.477 0.139749    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## NH4                   2.6303     0.3802    0.7955    8.6968
## dx_yr                 1.2049     0.8300    0.9410    1.5427
## age_dx                1.0036     0.9965    1.0000    1.0071
## sexF                  0.8669     1.1535    0.7327    1.0257
## dich_RaceNon-White    1.0159     0.9844    0.8834    1.1683
## smokeHxFormer         1.0995     0.9095    1.0521    1.1490
## smokeHxAlways         0.9817     1.0187    0.8074    1.1936
## smokeHxUnknown        1.3552     0.7379    0.9243    1.9870
## smokeHxEver           1.0355     0.9657    1.0268    1.0443
## disadv                1.0542     0.9486    0.8107    1.3709
## site02R               1.4024     0.7131    0.8574    2.2938
## site03R               0.7224     1.3842    0.6139    0.8502
## site04R               0.4498     2.2231    0.1986    1.0191
## site05R               0.3749     2.6674    0.1811    0.7762
## site06R               0.6658     1.5020    0.5416    0.8185
## site07R               0.4135     2.4182    0.1872    0.9135
## site09R               0.6641     1.5058    0.4683    0.9418
## site1                 0.5978     1.6728    0.3413    1.0472
## site101               0.5603     1.7849    0.2815    1.1150
## site102               0.6726     1.4868    0.5049    0.8960
## site103               1.0085     0.9915    0.9168    1.1095
## site104               0.7471     1.3385    0.6401    0.8719
## site105               0.7248     1.3798    0.3996    1.3146
## site106               0.4517     2.2138    0.2218    0.9199
## site107               0.7842     1.2753    0.4225    1.4554
## site108               0.8170     1.2241    0.4447    1.5010
## site10R               0.8203     1.2191    0.7769    0.8660
## site11R               0.7270     1.3755    0.5402    0.9784
## site12R               0.6487     1.5414    0.4869    0.8645
## site13R               0.4670     2.1414    0.2490    0.8757
## site14R               0.3897     2.5659    0.2483    0.6117
## site15R               0.8673     1.1531    0.7869    0.9558
## site16R               0.7952     1.2576    0.5869    1.0774
## site17R               0.6127     1.6322    0.3490    1.0756
## site18R               0.5824     1.7171    0.3612    0.9389
## site19R               0.9918     1.0083    0.8957    1.0982
## site20R               0.5323     1.8785    0.3579    0.7917
## site21R               0.5515     1.8133    0.4082    0.7451
## site22R               0.5429     1.8420    0.3005    0.9806
## site23R               0.5991     1.6692    0.4481    0.8010
## site24R               0.5854     1.7082    0.4311    0.7951
## site25R               0.7046     1.4193    0.4206    1.1801
## site26R               0.4740     2.1098    0.3012    0.7458
## site27R               0.5696     1.7557    0.2471    1.3127
## site28R               0.4436     2.2543    0.2156    0.9125
## site29R               0.5381     1.8585    0.3386    0.8551
## site30R               0.4029     2.4818    0.1705    0.9521
## site31R               0.5262     1.9003    0.3575    0.7745
## site32R               0.6321     1.5820    0.5531    0.7224
## site33R               0.4973     2.0110    0.3166    0.7811
## site34R               0.5801     1.7238    0.3580    0.9400
## site35R               0.7140     1.4005    0.5856    0.8706
## site36R               0.3960     2.5252    0.1827    0.8581
## site37R               0.4194     2.3841    0.1862    0.9448
## site38R               0.5832     1.7148    0.4599    0.7395
## site39R               0.6516     1.5347    0.5383    0.7887
## site40R               0.5199     1.9234    0.1772    1.5251
## site41R               0.5280     1.8941    0.3381    0.8245
## site42R               0.6422     1.5571    0.3568    1.1559
## 
## Concordance= 0.607  (se = 0.04 )
## Likelihood ratio test= 456.5  on 59 df,   p=<2e-16
## Wald test            = 2.62  on 59 df,   p=1
## Score (logrank) test = 432.6  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

91.5 IPF-Only BC

91.5.1 IPF-Only BC - Simmons Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ BC + dx_yr, data=Simm_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC + dx_yr, data = Simm_IPF, 
##     id = ID)
## 
##   n= 31982, number of events= 695 
##    (446 observations deleted due to missingness)
## 
##           coef exp(coef) se(coef)     z Pr(>|z|)   
## BC    0.628259  1.874345 0.210362 2.987  0.00282 **
## dx_yr 0.028760  1.029178 0.008759 3.283  0.00103 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##       exp(coef) exp(-coef) lower .95 upper .95
## BC        1.874     0.5335     1.241     2.831
## dx_yr     1.029     0.9716     1.012     1.047
## 
## Concordance= 0.526  (se = 0.013 )
## Likelihood ratio test= 16.39  on 2 df,   p=3e-04
## Wald test            = 16.41  on 2 df,   p=3e-04
## Score (logrank) test = 16.42  on 2 df,   p=3e-04

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ BC + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv, data=Simm_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv, data = Simm_IPF, id = ID)
## 
##   n= 30488, number of events= 670 
##    (1940 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## BC                  0.632992  1.883237  0.219097  2.889  0.00386 ** 
## dx_yr               0.016479  1.016616  0.009334  1.765  0.07749 .  
## age_dx              0.001827  1.001828  0.004544  0.402  0.68766    
## sexF               -0.369143  0.691327  0.088891 -4.153 3.29e-05 ***
## dich_RaceNon-White  0.156125  1.168973  0.128500  1.215  0.22437    
## smokeHxFormer       0.078484  1.081646  0.091931  0.854  0.39326    
## smokeHxAlways      -0.394385  0.674095  0.266559 -1.480  0.13900    
## smokeHxUnknown      0.468359  1.597371  0.170061  2.754  0.00589 ** 
## disadv              0.411600  1.509231  0.134154  3.068  0.00215 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## BC                    1.8832     0.5310    1.2258    2.8933
## dx_yr                 1.0166     0.9837    0.9982    1.0354
## age_dx                1.0018     0.9982    0.9929    1.0108
## sexF                  0.6913     1.4465    0.5808    0.8229
## dich_RaceNon-White    1.1690     0.8555    0.9087    1.5038
## smokeHxFormer         1.0816     0.9245    0.9033    1.2952
## smokeHxAlways         0.6741     1.4835    0.3998    1.1366
## smokeHxUnknown        1.5974     0.6260    1.1446    2.2293
## disadv                1.5092     0.6626    1.1603    1.9631
## 
## Concordance= 0.591  (se = 0.012 )
## Likelihood ratio test= 59.85  on 9 df,   p=1e-09
## Wald test            = 60.57  on 9 df,   p=1e-09
## Score (logrank) test = 61.4  on 9 df,   p=7e-10

91.5.2 IPF-Only BC - PFF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ BC + dx_yr + site, data=PFF_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC + dx_yr + site, 
##     data = PFF_IPF, id = ID)
## 
##   n= 54499, number of events= 1146 
##    (217 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef)      z Pr(>|z|)    
## BC       0.099257  1.104350  0.191189  0.519   0.6037    
## dx_yr    0.106755  1.112661  0.014418  7.404 1.32e-13 ***
## site02R  0.646203  1.908281  0.336503  1.920   0.0548 .  
## site03R -0.253290  0.776243  0.347464 -0.729   0.4660    
## site04R -0.226163  0.797588  0.451087 -0.501   0.6161    
## site05R -0.330905  0.718273  0.373927 -0.885   0.3762    
## site06R -0.043541  0.957393  0.327802 -0.133   0.8943    
## site07R -0.316679  0.728565  0.306205 -1.034   0.3010    
## site08R        NA        NA  0.000000     NA       NA    
## site09R -0.023325  0.976945  0.340386 -0.069   0.9454    
## site10R -0.134990  0.873725  0.385840 -0.350   0.7264    
## site11R -0.089347  0.914528  0.306042 -0.292   0.7703    
## site12R -0.075371  0.927399  0.334105 -0.226   0.8215    
## site13R -0.306567  0.735969  0.304772 -1.006   0.3145    
## site14R -0.371014  0.690034  0.642609 -0.577   0.5637    
## site15R  0.022444  1.022698  0.332948  0.067   0.9463    
## site16R  0.125508  1.133725  0.346084  0.363   0.7169    
## site17R  0.021720  1.021957  0.362292  0.060   0.9522    
## site18R -0.280953  0.755064  0.323999 -0.867   0.3859    
## site19R  0.003931  1.003938  0.350860  0.011   0.9911    
## site20R -0.250064  0.778751  0.347470 -0.720   0.4717    
## site21R -0.317079  0.728273  0.327771 -0.967   0.3334    
## site22R -0.079677  0.923414  0.309275 -0.258   0.7967    
## site23R -0.161592  0.850789  0.323082 -0.500   0.6170    
## site24R -0.132223  0.876145  0.331025 -0.399   0.6896    
## site25R -0.011326  0.988738  0.325267 -0.035   0.9722    
## site26R -0.394903  0.673745  0.344604 -1.146   0.2518    
## site27R  0.004993  1.005006  0.760579  0.007   0.9948    
## site28R -0.142427  0.867251  0.438201 -0.325   0.7452    
## site29R -0.280489  0.755414  0.433934 -0.646   0.5180    
## site30R -0.229097  0.795251  0.336184 -0.681   0.4956    
## site31R -0.322061  0.724654  0.450132 -0.715   0.4743    
## site32R -0.351340  0.703745  0.369009 -0.952   0.3410    
## site33R -0.334623  0.715608  0.332471 -1.006   0.3142    
## site34R -0.139899  0.869446  0.313538 -0.446   0.6555    
## site35R -0.105720  0.899676  0.333851 -0.317   0.7515    
## site36R -0.219538  0.802889  0.342673 -0.641   0.5217    
## site37R -0.183338  0.832486  0.337367 -0.543   0.5868    
## site38R -0.291973  0.746789  0.330256 -0.884   0.3767    
## site39R -0.210404  0.810257  0.360824 -0.583   0.5598    
## site40R  0.085412  1.089165  0.384113  0.222   0.8240    
## site41R -0.224165  0.799183  0.372044 -0.603   0.5468    
## site42R  0.064179  1.066283  0.362462  0.177   0.8595    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## BC         1.1043     0.9055    0.7592     1.606
## dx_yr      1.1127     0.8987    1.0817     1.145
## site02R    1.9083     0.5240    0.9868     3.690
## site03R    0.7762     1.2883    0.3929     1.534
## site04R    0.7976     1.2538    0.3295     1.931
## site05R    0.7183     1.3922    0.3451     1.495
## site06R    0.9574     1.0445    0.5036     1.820
## site07R    0.7286     1.3726    0.3998     1.328
## site08R        NA         NA        NA        NA
## site09R    0.9769     1.0236    0.5013     1.904
## site10R    0.8737     1.1445    0.4102     1.861
## site11R    0.9145     1.0935    0.5020     1.666
## site12R    0.9274     1.0783    0.4818     1.785
## site13R    0.7360     1.3588    0.4050     1.337
## site14R    0.6900     1.4492    0.1958     2.431
## site15R    1.0227     0.9778    0.5325     1.964
## site16R    1.1337     0.8820    0.5753     2.234
## site17R    1.0220     0.9785    0.5024     2.079
## site18R    0.7551     1.3244    0.4001     1.425
## site19R    1.0039     0.9961    0.5047     1.997
## site20R    0.7788     1.2841    0.3941     1.539
## site21R    0.7283     1.3731    0.3831     1.385
## site22R    0.9234     1.0829    0.5037     1.693
## site23R    0.8508     1.1754    0.4517     1.603
## site24R    0.8761     1.1414    0.4579     1.676
## site25R    0.9887     1.0114    0.5227     1.870
## site26R    0.6737     1.4842    0.3429     1.324
## site27R    1.0050     0.9950    0.2263     4.462
## site28R    0.8673     1.1531    0.3674     2.047
## site29R    0.7554     1.3238    0.3227     1.768
## site30R    0.7953     1.2575    0.4115     1.537
## site31R    0.7247     1.3800    0.2999     1.751
## site32R    0.7037     1.4210    0.3414     1.450
## site33R    0.7156     1.3974    0.3730     1.373
## site34R    0.8694     1.1502    0.4703     1.607
## site35R    0.8997     1.1115    0.4676     1.731
## site36R    0.8029     1.2455    0.4102     1.572
## site37R    0.8325     1.2012    0.4297     1.613
## site38R    0.7468     1.3391    0.3909     1.427
## site39R    0.8103     1.2342    0.3995     1.643
## site40R    1.0892     0.9181    0.5130     2.312
## site41R    0.7992     1.2513    0.3854     1.657
## site42R    1.0663     0.9378    0.5240     2.170
## 
## Concordance= 0.59  (se = 0.009 )
## Likelihood ratio test= 110.8  on 42 df,   p=4e-08
## Wald test            = 109.1  on 42 df,   p=7e-08
## Score (logrank) test = 113.3  on 42 df,   p=2e-08

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ BC + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=PFF_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = PFF_IPF, 
##     id = ID)
## 
##   n= 53867, number of events= 1131 
##    (849 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)      z Pr(>|z|)    
## BC                  0.1218474  1.1295818  0.1955235  0.623   0.5332    
## dx_yr               0.1061118  1.1119462  0.0149235  7.110 1.16e-12 ***
## age_dx             -0.0003375  0.9996625  0.0039364 -0.086   0.9317    
## sexM                0.0616244  1.0635628  0.0712850  0.864   0.3873    
## dich_RaceNon-White -0.0620134  0.9398703  0.1261843 -0.491   0.6231    
## smokeHxEver         0.0429746  1.0439114  0.0646559  0.665   0.5063    
## disadv             -0.0578564  0.9437854  0.1124418 -0.515   0.6069    
## site02R             0.5909210  1.8056506  0.3495292  1.691   0.0909 .  
## site03R            -0.2872648  0.7503131  0.3602265 -0.797   0.4252    
## site04R            -0.2878085  0.7499052  0.4624276 -0.622   0.5337    
## site05R            -0.3840169  0.6811199  0.3877908 -0.990   0.3220    
## site06R            -0.0857503  0.9178234  0.3443273 -0.249   0.8033    
## site07R            -0.3646471  0.6944417  0.3210754 -1.136   0.2561    
## site08R                    NA         NA  0.0000000     NA       NA    
## site09R            -0.1088370  0.8968766  0.3555378 -0.306   0.7595    
## site10R            -0.1912467  0.8259288  0.3987437 -0.480   0.6315    
## site11R            -0.1551920  0.8562508  0.3257545 -0.476   0.6338    
## site12R            -0.1205837  0.8864029  0.3476512 -0.347   0.7287    
## site13R            -0.3708642  0.6901376  0.3194745 -1.161   0.2457    
## site14R            -0.4411488  0.6432970  0.6508203 -0.678   0.4979    
## site15R            -0.0316308  0.9688642  0.3494681 -0.091   0.9279    
## site16R             0.0743534  1.0771874  0.3584692  0.207   0.8357    
## site17R            -0.0420303  0.9588407  0.3767215 -0.112   0.9112    
## site18R            -0.3196326  0.7264159  0.3400481 -0.940   0.3472    
## site19R            -0.0418402  0.9590230  0.3627934 -0.115   0.9082    
## site20R            -0.3058402  0.7365043  0.3616359 -0.846   0.3977    
## site21R            -0.3700646  0.6906897  0.3433973 -1.078   0.2812    
## site22R            -0.1409431  0.8685387  0.3251488 -0.433   0.6647    
## site23R            -0.2292489  0.7951306  0.3392161 -0.676   0.4992    
## site24R            -0.1899407  0.8270082  0.3445719 -0.551   0.5815    
## site25R            -0.0462749  0.9547795  0.3385172 -0.137   0.8913    
## site26R            -0.4296493  0.6507373  0.3575209 -1.202   0.2295    
## site27R            -0.1097219  0.8960833  0.7722682 -0.142   0.8870    
## site28R            -0.1818867  0.8336958  0.4508562 -0.403   0.6866    
## site29R            -0.3386558  0.7127277  0.4456247 -0.760   0.4473    
## site30R            -0.2920598  0.7467239  0.3507292 -0.833   0.4050    
## site31R            -0.3775256  0.6855556  0.4597307 -0.821   0.4115    
## site32R            -0.4175668  0.6586475  0.3917752 -1.066   0.2865    
## site33R            -0.3768800  0.6859984  0.3481573 -1.082   0.2790    
## site34R            -0.1856758  0.8305428  0.3289784 -0.564   0.5725    
## site35R            -0.1691685  0.8443666  0.3483704 -0.486   0.6273    
## site36R            -0.2596634  0.7713112  0.3584497 -0.724   0.4688    
## site37R            -0.2340428  0.7913279  0.3514456 -0.666   0.5054    
## site38R            -0.3344544  0.7157285  0.3445620 -0.971   0.3317    
## site39R            -0.2666567  0.7659360  0.3740019 -0.713   0.4759    
## site40R             0.0332337  1.0337921  0.3995037  0.083   0.9337    
## site41R            -0.2701151  0.7632916  0.3836329 -0.704   0.4814    
## site42R            -0.0092087  0.9908336  0.3759002 -0.024   0.9805    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## BC                    1.1296     0.8853    0.7700     1.657
## dx_yr                 1.1119     0.8993    1.0799     1.145
## age_dx                0.9997     1.0003    0.9920     1.007
## sexM                  1.0636     0.9402    0.9249     1.223
## dich_RaceNon-White    0.9399     1.0640    0.7339     1.204
## smokeHxEver           1.0439     0.9579    0.9197     1.185
## disadv                0.9438     1.0596    0.7571     1.176
## site02R               1.8057     0.5538    0.9102     3.582
## site03R               0.7503     1.3328    0.3704     1.520
## site04R               0.7499     1.3335    0.3030     1.856
## site05R               0.6811     1.4682    0.3185     1.457
## site06R               0.9178     1.0895    0.4674     1.802
## site07R               0.6944     1.4400    0.3701     1.303
## site08R                   NA         NA        NA        NA
## site09R               0.8969     1.1150    0.4468     1.800
## site10R               0.8259     1.2108    0.3780     1.804
## site11R               0.8563     1.1679    0.4522     1.621
## site12R               0.8864     1.1282    0.4484     1.752
## site13R               0.6901     1.4490    0.3690     1.291
## site14R               0.6433     1.5545    0.1797     2.304
## site15R               0.9689     1.0321    0.4884     1.922
## site16R               1.0772     0.9283    0.5335     2.175
## site17R               0.9588     1.0429    0.4582     2.006
## site18R               0.7264     1.3766    0.3730     1.415
## site19R               0.9590     1.0427    0.4710     1.953
## site20R               0.7365     1.3578    0.3625     1.496
## site21R               0.6907     1.4478    0.3524     1.354
## site22R               0.8685     1.1514    0.4592     1.643
## site23R               0.7951     1.2577    0.4090     1.546
## site24R               0.8270     1.2092    0.4209     1.625
## site25R               0.9548     1.0474    0.4918     1.854
## site26R               0.6507     1.5367    0.3229     1.311
## site27R               0.8961     1.1160    0.1972     4.071
## site28R               0.8337     1.1995    0.3445     2.017
## site29R               0.7127     1.4031    0.2976     1.707
## site30R               0.7467     1.3392    0.3755     1.485
## site31R               0.6856     1.4587    0.2784     1.688
## site32R               0.6586     1.5183    0.3056     1.419
## site33R               0.6860     1.4577    0.3467     1.357
## site34R               0.8305     1.2040    0.4358     1.583
## site35R               0.8444     1.1843    0.4266     1.671
## site36R               0.7713     1.2965    0.3820     1.557
## site37R               0.7913     1.2637    0.3974     1.576
## site38R               0.7157     1.3972    0.3643     1.406
## site39R               0.7659     1.3056    0.3680     1.594
## site40R               1.0338     0.9673    0.4725     2.262
## site41R               0.7633     1.3101    0.3599     1.619
## site42R               0.9908     1.0093    0.4743     2.070
## 
## Concordance= 0.592  (se = 0.009 )
## Likelihood ratio test= 111.6  on 47 df,   p=4e-07
## Wald test            = 110.2  on 47 df,   p=5e-07
## Score (logrank) test = 114.5  on 47 df,   p=1e-07

91.5.3 IPF-Only BC - CARE-PF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ BC + dx_yr + site, data=CARE_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC + dx_yr + site, 
##     data = CARE_IPF, id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef)      z Pr(>|z|)    
## BC      -0.22373   0.79953  0.21147 -1.058  0.29007    
## dx_yr    0.81648   2.26252  0.03239 25.209  < 2e-16 ***
## site102  0.19599   1.21651  0.14898  1.315  0.18835    
## site103  0.50566   1.65808  0.12013  4.209 2.56e-05 ***
## site104  0.36060   1.43418  0.13978  2.580  0.00989 ** 
## site105  0.05329   1.05474  0.12653  0.421  0.67363    
## site106  0.14127   1.15173  0.12046  1.173  0.24088    
## site107  0.09373   1.09827  0.19780  0.474  0.63559    
## site108 -0.43347   0.64826  0.26575 -1.631  0.10286    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## BC         0.7995     1.2507    0.5282     1.210
## dx_yr      2.2625     0.4420    2.1234     2.411
## site102    1.2165     0.8220    0.9084     1.629
## site103    1.6581     0.6031    1.3103     2.098
## site104    1.4342     0.6973    1.0905     1.886
## site105    1.0547     0.9481    0.8231     1.352
## site106    1.1517     0.8683    0.9095     1.458
## site107    1.0983     0.9105    0.7453     1.618
## site108    0.6483     1.5426    0.3851     1.091
## 
## Concordance= 0.774  (se = 0.01 )
## Likelihood ratio test= 1053  on 9 df,   p=<2e-16
## Wald test            = 665.1  on 9 df,   p=<2e-16
## Score (logrank) test = 588.6  on 9 df,   p=<2e-16

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ BC + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=CARE_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = CARE_IPF, 
##     id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## BC                 -0.271875  0.761950  0.216355 -1.257  0.20889    
## dx_yr               0.816520  2.262612  0.032570 25.070  < 2e-16 ***
## age_dx              0.012385  1.012462  0.004511  2.745  0.00605 ** 
## sexF               -0.043440  0.957490  0.077028 -0.564  0.57279    
## dich_RaceNon-White -0.139022  0.870209  0.110288 -1.261  0.20748    
## smokeHxFormer       0.002227  1.002229  0.081503  0.027  0.97820    
## smokeHxAlways       0.051606  1.052961  0.168804  0.306  0.75982    
## smokeHxUnknown      0.406790  1.501988  0.734622  0.554  0.57976    
## disadv              0.001681  1.001683  0.130506  0.013  0.98972    
## site102             0.173283  1.189203  0.151316  1.145  0.25214    
## site103             0.517544  1.677902  0.123027  4.207 2.59e-05 ***
## site104             0.366238  1.442298  0.140536  2.606  0.00916 ** 
## site105             0.069771  1.072263  0.129673  0.538  0.59054    
## site106             0.113163  1.119815  0.121638  0.930  0.35220    
## site107             0.028870  1.029291  0.205325  0.141  0.88818    
## site108            -0.440094  0.643976  0.269385 -1.634  0.10232    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## BC                    0.7619     1.3124    0.4986     1.164
## dx_yr                 2.2626     0.4420    2.1227     2.412
## age_dx                1.0125     0.9877    1.0035     1.021
## sexF                  0.9575     1.0444    0.8233     1.114
## dich_RaceNon-White    0.8702     1.1491    0.7010     1.080
## smokeHxFormer         1.0022     0.9978    0.8543     1.176
## smokeHxAlways         1.0530     0.9497    0.7564     1.466
## smokeHxUnknown        1.5020     0.6658    0.3559     6.338
## disadv                1.0017     0.9983    0.7756     1.294
## site102               1.1892     0.8409    0.8840     1.600
## site103               1.6779     0.5960    1.3184     2.135
## site104               1.4423     0.6933    1.0950     1.900
## site105               1.0723     0.9326    0.8316     1.383
## site106               1.1198     0.8930    0.8823     1.421
## site107               1.0293     0.9715    0.6883     1.539
## site108               0.6440     1.5529    0.3798     1.092
## 
## Concordance= 0.776  (se = 0.01 )
## Likelihood ratio test= 1065  on 16 df,   p=<2e-16
## Wald test            = 672  on 16 df,   p=<2e-16
## Score (logrank) test = 596.7  on 16 df,   p=<2e-16

91.5.4 IPF-Only BC - Combined Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ BC + dx_yr + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC + dx_yr + site, 
##     data = All_IPF, id = ID, cluster = cohort)
## 
##   n= 123367, number of events= 2749 
##    (723 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## BC       0.408909  1.505174  0.116883  0.159890   2.557 0.010545 *  
## dx_yr    0.109609  1.115842  0.007103  0.076699   1.429 0.152979    
## site02R  0.625110  1.868451  0.334563  0.078830   7.930 2.19e-15 ***
## site03R -0.312166  0.731860  0.347266  0.036797  -8.483  < 2e-16 ***
## site04R -0.299877  0.740909  0.449878  0.097508  -3.075 0.002102 ** 
## site05R -0.462887  0.629464  0.370487  0.070489  -6.567 5.14e-11 ***
## site06R -0.138325  0.870815  0.325963  0.081221  -1.703 0.088553 .  
## site07R -0.376194  0.686469  0.305101  0.109113  -3.448 0.000565 ***
## site09R -0.081960  0.921309  0.339988  0.032784  -2.500 0.012419 *  
## site1    0.117649  1.124849  0.283576  0.107991   1.089 0.275962    
## site101 -0.130442  0.877708  0.293320  0.140283  -0.930 0.352451    
## site102 -0.207992  0.812213  0.298007  0.101098  -2.057 0.039655 *  
## site103  0.119517  1.126952  0.286364  0.080669   1.482 0.138454    
## site104 -0.174801  0.839624  0.294721  0.083322  -2.098 0.035914 *  
## site105  0.082066  1.085528  0.289585  0.141993   0.578 0.563293    
## site106 -0.265428  0.766878  0.286713  0.103167  -2.573 0.010088 *  
## site107  0.215669  1.240691  0.326760  0.154675   1.394 0.163217    
## site108  0.188009  1.206845  0.369696  0.196000   0.959 0.337444    
## site10R -0.103499  0.901677  0.385491  0.057116  -1.812 0.069972 .  
## site11R -0.190240  0.826761  0.301585  0.087523  -2.174 0.029736 *  
## site12R -0.084005  0.919427  0.333912  0.029152  -2.882 0.003957 ** 
## site13R -0.379882  0.683942  0.302630  0.111507  -3.407 0.000657 ***
## site14R -0.468043  0.626227  0.641402  0.126051  -3.713 0.000205 ***
## site15R -0.050746  0.950520  0.331279  0.038778  -1.309 0.190667    
## site16R  0.065525  1.067719  0.345034  0.041103   1.594 0.110903    
## site17R -0.045170  0.955835  0.360760  0.034077  -1.326 0.184992    
## site18R -0.317852  0.727711  0.321967  0.161086  -1.973 0.048475 *  
## site19R  0.021805  1.022044  0.350239  0.009115   2.392 0.016749 *  
## site20R -0.303856  0.737967  0.347186  0.030297 -10.029  < 2e-16 ***
## site21R -0.467824  0.626363  0.322356  0.094747  -4.938 7.91e-07 ***
## site22R -0.179265  0.835884  0.306645  0.052928  -3.387 0.000707 ***
## site23R -0.207165  0.812885  0.322648  0.028442  -7.284 3.25e-13 ***
## site24R -0.160489  0.851727  0.330657  0.055043  -2.916 0.003549 ** 
## site25R -0.049299  0.951896  0.324959  0.050176  -0.983 0.325844    
## site26R -0.415223  0.660193  0.344468  0.077119  -5.384 7.28e-08 ***
## site27R -0.083809  0.919607  0.759895  0.125006  -0.670 0.502580    
## site28R -0.276293  0.758590  0.435366  0.068437  -4.037 5.41e-05 ***
## site29R -0.332769  0.716936  0.433757  0.050387  -6.604 4.00e-11 ***
## site30R -0.334010  0.716047  0.332289  0.126003  -2.651 0.008030 ** 
## site31R -0.384518  0.680779  0.449616  0.075222  -5.112 3.19e-07 ***
## site32R -0.395489  0.673351  0.368694  0.040058  -9.873  < 2e-16 ***
## site33R -0.371889  0.689431  0.332191  0.085997  -4.324 1.53e-05 ***
## site34R -0.170611  0.843149  0.313449  0.041747  -4.087 4.37e-05 ***
## site35R -0.122969  0.884291  0.333805  0.027042  -4.547 5.43e-06 ***
## site36R -0.344904  0.708289  0.339454  0.065747  -5.246 1.56e-07 ***
## site37R -0.286127  0.751167  0.335069  0.073930  -3.870 0.000109 ***
## site38R -0.358490  0.698731  0.329367  0.048223  -7.434 1.05e-13 ***
## site39R -0.275219  0.759406  0.360436  0.039343  -6.995 2.65e-12 ***
## site40R -0.200190  0.818575  0.359432  0.206760  -0.968 0.332930    
## site41R -0.321460  0.725090  0.369764  0.067069  -4.793 1.64e-06 ***
## site42R -0.004633  0.995378  0.360765  0.046262  -0.100 0.920229    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## BC         1.5052     0.6644    1.1002    2.0591
## dx_yr      1.1158     0.8962    0.9601    1.2968
## site02R    1.8685     0.5352    1.6010    2.1806
## site03R    0.7319     1.3664    0.6809    0.7866
## site04R    0.7409     1.3497    0.6120    0.8969
## site05R    0.6295     1.5887    0.5482    0.7227
## site06R    0.8708     1.1483    0.7427    1.0211
## site07R    0.6865     1.4567    0.5543    0.8502
## site09R    0.9213     1.0854    0.8640    0.9825
## site1      1.1248     0.8890    0.9103    1.3900
## site101    0.8777     1.1393    0.6667    1.1555
## site102    0.8122     1.2312    0.6662    0.9902
## site103    1.1270     0.8873    0.9621    1.3200
## site104    0.8396     1.1910    0.7131    0.9886
## site105    1.0855     0.9212    0.8218    1.4339
## site106    0.7669     1.3040    0.6265    0.9387
## site107    1.2407     0.8060    0.9162    1.6801
## site108    1.2068     0.8286    0.8219    1.7721
## site10R    0.9017     1.1090    0.8062    1.0085
## site11R    0.8268     1.2095    0.6964    0.9815
## site12R    0.9194     1.0876    0.8684    0.9735
## site13R    0.6839     1.4621    0.5497    0.8510
## site14R    0.6262     1.5969    0.4891    0.8017
## site15R    0.9505     1.0521    0.8810    1.0256
## site16R    1.0677     0.9366    0.9851    1.1573
## site17R    0.9558     1.0462    0.8941    1.0219
## site18R    0.7277     1.3742    0.5307    0.9979
## site19R    1.0220     0.9784    1.0039    1.0405
## site20R    0.7380     1.3551    0.6954    0.7831
## site21R    0.6264     1.5965    0.5202    0.7542
## site22R    0.8359     1.1963    0.7535    0.9273
## site23R    0.8129     1.2302    0.7688    0.8595
## site24R    0.8517     1.1741    0.7646    0.9488
## site25R    0.9519     1.0505    0.8627    1.0503
## site26R    0.6602     1.5147    0.5676    0.7679
## site27R    0.9196     1.0874    0.7198    1.1749
## site28R    0.7586     1.3182    0.6634    0.8675
## site29R    0.7169     1.3948    0.6495    0.7914
## site30R    0.7160     1.3966    0.5594    0.9166
## site31R    0.6808     1.4689    0.5875    0.7889
## site32R    0.6734     1.4851    0.6225    0.7283
## site33R    0.6894     1.4505    0.5825    0.8160
## site34R    0.8431     1.1860    0.7769    0.9150
## site35R    0.8843     1.1308    0.8386    0.9324
## site36R    0.7083     1.4119    0.6227    0.8057
## site37R    0.7512     1.3313    0.6498    0.8683
## site38R    0.6987     1.4312    0.6357    0.7680
## site39R    0.7594     1.3168    0.7030    0.8203
## site40R    0.8186     1.2216    0.5458    1.2276
## site41R    0.7251     1.3791    0.6358    0.8270
## site42R    0.9954     1.0046    0.9091    1.0898
## 
## Concordance= 0.596  (se = 0.042 )
## Likelihood ratio test= 360  on 51 df,   p=<2e-16
## Wald test            = 6.69  on 51 df,   p=1
## Score (logrank) test = 349.8  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ BC + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ BC + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = All_IPF, 
##     id = ID, cluster = cohort)
## 
##   n= 121241, number of events= 2709 
##    (2849 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## BC                  0.408936  1.505215  0.119460  0.168224   2.431 0.015061 *  
## dx_yr               0.106752  1.112659  0.007493  0.082569   1.293 0.196047    
## age_dx              0.003889  1.003896  0.002446  0.001895   2.052 0.040125 *  
## sexF               -0.154513  0.856832  0.044855  0.082701  -1.868 0.061714 .  
## dich_RaceNon-White  0.012831  1.012914  0.068754  0.062744   0.204 0.837965    
## smokeHxFormer       0.101466  1.106792  0.059347  0.032842   3.089 0.002005 ** 
## smokeHxAlways      -0.029455  0.970975  0.138344  0.109023  -0.270 0.787027    
## smokeHxUnknown      0.284492  1.329086  0.154015  0.241709   1.177 0.239196    
## smokeHxEver         0.036456  1.037129  0.064320  0.008009   4.552 5.32e-06 ***
## disadv              0.025355  1.025679  0.070567  0.143259   0.177 0.859520    
## site02R             0.590435  1.804773  0.345515  0.107672   5.484 4.17e-08 ***
## site03R            -0.311144  0.732609  0.357692  0.043370  -7.174 7.27e-13 ***
## site04R            -0.316361  0.728797  0.458543  0.132193  -2.393 0.016704 *  
## site05R            -0.478389  0.619781  0.381622  0.067784  -7.058 1.69e-12 ***
## site06R            -0.165352  0.847595  0.339043  0.088767  -1.863 0.062494 .  
## site07R            -0.402471  0.668666  0.317079  0.120541  -3.339 0.000841 ***
## site09R            -0.150497  0.860280  0.352699  0.031104  -4.839 1.31e-06 ***
## site1               0.031751  1.032261  0.300532  0.117353   0.271 0.786727    
## site101            -0.202370  0.816793  0.308489  0.139190  -1.454 0.145970    
## site102            -0.299791  0.740973  0.313616  0.109857  -2.729 0.006354 ** 
## site103             0.025321  1.025644  0.301884  0.053204   0.476 0.634134    
## site104            -0.258297  0.772366  0.310474  0.062390  -4.140 3.47e-05 ***
## site105             0.002697  1.002701  0.306691  0.137500   0.020 0.984348    
## site106            -0.364439  0.694586  0.302984  0.100196  -3.637 0.000276 ***
## site107             0.105752  1.111546  0.342348  0.137555   0.769 0.442015    
## site108             0.095608  1.100328  0.381919  0.179467   0.533 0.594217    
## site10R            -0.183664  0.832216  0.396191  0.031993  -5.741 9.42e-09 ***
## site11R            -0.241137  0.785734  0.316600  0.091484  -2.636 0.008393 ** 
## site12R            -0.122088  0.885070  0.345083  0.029585  -4.127 3.68e-05 ***
## site13R            -0.437616  0.645574  0.314763  0.130048  -3.365 0.000765 ***
## site14R            -0.499207  0.607012  0.647570  0.070585  -7.072 1.52e-12 ***
## site15R            -0.105749  0.899651  0.343850  0.036300  -2.913 0.003577 ** 
## site16R             0.046339  1.047429  0.355574  0.022563   2.054 0.039995 *  
## site17R            -0.076516  0.926338  0.372000  0.034609  -2.211 0.027043 *  
## site18R            -0.363720  0.695086  0.334228  0.146910  -2.476 0.013293 *  
## site19R            -0.038922  0.961826  0.360840  0.019815  -1.964 0.049500 *  
## site20R            -0.321442  0.725103  0.358697  0.028304 -11.357  < 2e-16 ***
## site21R            -0.521575  0.593585  0.334488  0.099838  -5.224 1.75e-07 ***
## site22R            -0.210279  0.810358  0.319251  0.060829  -3.457 0.000546 ***
## site23R            -0.282112  0.754189  0.336628  0.027822 -10.140  < 2e-16 ***
## site24R            -0.192360  0.825010  0.342350  0.048542  -3.963 7.41e-05 ***
## site25R            -0.081533  0.921702  0.336947  0.077658  -1.050 0.293762    
## site26R            -0.469183  0.625513  0.355377  0.057682  -8.134 4.16e-16 ***
## site27R            -0.097607  0.907005  0.767527  0.158219  -0.617 0.537294    
## site28R            -0.292598  0.746322  0.444601  0.059515  -4.916 8.82e-07 ***
## site29R            -0.359068  0.698327  0.443154  0.092760  -3.871 0.000108 ***
## site30R            -0.415753  0.659843  0.345074  0.138295  -3.006 0.002645 ** 
## site31R            -0.438249  0.645165  0.458039  0.080206  -5.464 4.65e-08 ***
## site32R            -0.468130  0.626172  0.389383  0.047394  -9.877  < 2e-16 ***
## site33R            -0.422899  0.655145  0.345442  0.067293  -6.284 3.29e-10 ***
## site34R            -0.203630  0.815764  0.326262  0.051157  -3.980 6.88e-05 ***
## site35R            -0.193826  0.823801  0.347008  0.027035  -7.169 7.54e-13 ***
## site36R            -0.367252  0.692635  0.351227  0.061994  -5.924 3.14e-09 ***
## site37R            -0.320615  0.725703  0.346459  0.090744  -3.533 0.000411 ***
## site38R            -0.382264  0.682315  0.340950  0.039546  -9.666  < 2e-16 ***
## site39R            -0.336650  0.714159  0.371390  0.042272  -7.964 1.67e-15 ***
## site40R            -0.258723  0.772037  0.371214  0.220692  -1.172 0.241065    
## site41R            -0.377377  0.685657  0.379526  0.074499  -5.066 4.07e-07 ***
## site42R            -0.050760  0.950507  0.371424  0.063214  -0.803 0.421990    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## BC                    1.5052     0.6644    1.0824    2.0931
## dx_yr                 1.1127     0.8987    0.9464    1.3081
## age_dx                1.0039     0.9961    1.0002    1.0076
## sexF                  0.8568     1.1671    0.7286    1.0076
## dich_RaceNon-White    1.0129     0.9873    0.8957    1.1455
## smokeHxFormer         1.1068     0.9035    1.0378    1.1804
## smokeHxAlways         0.9710     1.0299    0.7842    1.2023
## smokeHxUnknown        1.3291     0.7524    0.8276    2.1345
## smokeHxEver           1.0371     0.9642    1.0210    1.0535
## disadv                1.0257     0.9750    0.7746    1.3582
## site02R               1.8048     0.5541    1.4614    2.2288
## site03R               0.7326     1.3650    0.6729    0.7976
## site04R               0.7288     1.3721    0.5624    0.9443
## site05R               0.6198     1.6135    0.5427    0.7078
## site06R               0.8476     1.1798    0.7122    1.0087
## site07R               0.6687     1.4955    0.5280    0.8469
## site09R               0.8603     1.1624    0.8094    0.9144
## site1                 1.0323     0.9687    0.8202    1.2992
## site101               0.8168     1.2243    0.6218    1.0730
## site102               0.7410     1.3496    0.5974    0.9190
## site103               1.0256     0.9750    0.9241    1.1384
## site104               0.7724     1.2947    0.6835    0.8728
## site105               1.0027     0.9973    0.7658    1.3128
## site106               0.6946     1.4397    0.5707    0.8453
## site107               1.1115     0.8996    0.8489    1.4555
## site108               1.1003     0.9088    0.7740    1.5642
## site10R               0.8322     1.2016    0.7816    0.8861
## site11R               0.7857     1.2727    0.6568    0.9400
## site12R               0.8851     1.1299    0.8352    0.9379
## site13R               0.6456     1.5490    0.5003    0.8330
## site14R               0.6070     1.6474    0.5286    0.6971
## site15R               0.8997     1.1115    0.8379    0.9660
## site16R               1.0474     0.9547    1.0021    1.0948
## site17R               0.9263     1.0795    0.8656    0.9914
## site18R               0.6951     1.4387    0.5212    0.9270
## site19R               0.9618     1.0397    0.9252    0.9999
## site20R               0.7251     1.3791    0.6860    0.7665
## site21R               0.5936     1.6847    0.4881    0.7219
## site22R               0.8104     1.2340    0.7193    0.9130
## site23R               0.7542     1.3259    0.7142    0.7965
## site24R               0.8250     1.2121    0.7501    0.9074
## site25R               0.9217     1.0849    0.7916    1.0732
## site26R               0.6255     1.5987    0.5586    0.7004
## site27R               0.9070     1.1025    0.6652    1.2368
## site28R               0.7463     1.3399    0.6642    0.8387
## site29R               0.6983     1.4320    0.5822    0.8376
## site30R               0.6598     1.5155    0.5032    0.8653
## site31R               0.6452     1.5500    0.5513    0.7550
## site32R               0.6262     1.5970    0.5706    0.6871
## site33R               0.6551     1.5264    0.5742    0.7475
## site34R               0.8158     1.2258    0.7379    0.9018
## site35R               0.8238     1.2139    0.7813    0.8686
## site36R               0.6926     1.4438    0.6134    0.7821
## site37R               0.7257     1.3780    0.6075    0.8670
## site38R               0.6823     1.4656    0.6314    0.7373
## site39R               0.7142     1.4002    0.6574    0.7758
## site40R               0.7720     1.2953    0.5009    1.1898
## site41R               0.6857     1.4585    0.5925    0.7935
## site42R               0.9505     1.0521    0.8397    1.0759
## 
## Concordance= 0.6  (se = 0.039 )
## Likelihood ratio test= 391.5  on 59 df,   p=<2e-16
## Wald test            = 5.99  on 59 df,   p=1
## Score (logrank) test = 386.8  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

91.6 IPF-Only OM

91.6.1 IPF-Only OM - Simmons Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ OM + dx_yr, data=Simm_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM + dx_yr, data = Simm_IPF, 
##     id = ID)
## 
##   n= 31982, number of events= 695 
##    (446 observations deleted due to missingness)
## 
##           coef exp(coef) se(coef)     z Pr(>|z|)    
## OM    0.178291  1.195173 0.053759 3.316 0.000912 ***
## dx_yr 0.028757  1.029174 0.008606 3.342 0.000833 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##       exp(coef) exp(-coef) lower .95 upper .95
## OM        1.195     0.8367     1.076     1.328
## dx_yr     1.029     0.9717     1.012     1.047
## 
## Concordance= 0.541  (se = 0.013 )
## Likelihood ratio test= 18.24  on 2 df,   p=1e-04
## Wald test            = 18.76  on 2 df,   p=8e-05
## Score (logrank) test = 18.79  on 2 df,   p=8e-05

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ OM + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv, data=Simm_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv, data = Simm_IPF, id = ID)
## 
##   n= 30488, number of events= 670 
##    (1940 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## OM                  0.168922  1.184027  0.056134  3.009  0.00262 ** 
## dx_yr               0.016027  1.016156  0.009194  1.743  0.08131 .  
## age_dx              0.002990  1.002995  0.004558  0.656  0.51185    
## sexF               -0.365703  0.693708  0.088955 -4.111 3.94e-05 ***
## dich_RaceNon-White  0.137471  1.147369  0.128716  1.068  0.28551    
## smokeHxFormer       0.077953  1.081072  0.091881  0.848  0.39621    
## smokeHxAlways      -0.441039  0.643368  0.266488 -1.655  0.09792 .  
## smokeHxUnknown      0.454587  1.575522  0.170628  2.664  0.00772 ** 
## disadv              0.390925  1.478348  0.133259  2.934  0.00335 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## OM                    1.1840     0.8446    1.0607    1.3217
## dx_yr                 1.0162     0.9841    0.9980    1.0346
## age_dx                1.0030     0.9970    0.9941    1.0120
## sexF                  0.6937     1.4415    0.5827    0.8258
## dich_RaceNon-White    1.1474     0.8716    0.8915    1.4766
## smokeHxFormer         1.0811     0.9250    0.9029    1.2944
## smokeHxAlways         0.6434     1.5543    0.3816    1.0847
## smokeHxUnknown        1.5755     0.6347    1.1277    2.2012
## disadv                1.4783     0.6764    1.1385    1.9196
## 
## Concordance= 0.596  (se = 0.012 )
## Likelihood ratio test= 60.46  on 9 df,   p=1e-09
## Wald test            = 61.74  on 9 df,   p=6e-10
## Score (logrank) test = 62.65  on 9 df,   p=4e-10

91.6.2 IPF-Only OM - PFF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ OM + dx_yr + site, data=PFF_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM + dx_yr + site, 
##     data = PFF_IPF, id = ID)
## 
##   n= 54499, number of events= 1146 
##    (217 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef)      z Pr(>|z|)    
## OM       0.026811  1.027173  0.040359  0.664   0.5065    
## dx_yr    0.105067  1.110785  0.013605  7.723 1.14e-14 ***
## site02R  0.644361  1.904770  0.336296  1.916   0.0554 .  
## site03R -0.249479  0.779207  0.347173 -0.719   0.4724    
## site04R -0.202282  0.816865  0.449609 -0.450   0.6528    
## site05R -0.320918  0.725483  0.370195 -0.867   0.3860    
## site06R -0.034171  0.966407  0.325679 -0.105   0.9164    
## site07R -0.306329  0.736145  0.305021 -1.004   0.3152    
## site08R        NA        NA  0.000000     NA       NA    
## site09R -0.012780  0.987301  0.339751 -0.038   0.9700    
## site10R -0.115701  0.890742  0.387975 -0.298   0.7655    
## site11R -0.089278  0.914591  0.303396 -0.294   0.7686    
## site12R -0.061396  0.940451  0.335225 -0.183   0.8547    
## site13R -0.287192  0.750368  0.302020 -0.951   0.3417    
## site14R -0.344690  0.708440  0.641282 -0.538   0.5909    
## site15R  0.029671  1.030116  0.331072  0.090   0.9286    
## site16R  0.143857  1.154719  0.344433  0.418   0.6762    
## site17R  0.031972  1.032489  0.360320  0.089   0.9293    
## site18R -0.277546  0.757641  0.322995 -0.859   0.3902    
## site19R  0.019901  1.020100  0.352746  0.056   0.9550    
## site20R -0.238660  0.787683  0.347272 -0.687   0.4919    
## site21R -0.305290  0.736910  0.321809 -0.949   0.3428    
## site22R -0.052924  0.948452  0.305140 -0.173   0.8623    
## site23R -0.156379  0.855235  0.322545 -0.485   0.6278    
## site24R -0.118150  0.888563  0.331982 -0.356   0.7219    
## site25R -0.007785  0.992245  0.325029 -0.024   0.9809    
## site26R -0.382687  0.682027  0.345011 -1.109   0.2673    
## site27R  0.028110  1.028509  0.759854  0.037   0.9705    
## site28R -0.127422  0.880362  0.434516 -0.293   0.7693    
## site29R -0.259591  0.771367  0.434311 -0.598   0.5500    
## site30R -0.202961  0.816310  0.330850 -0.613   0.5396    
## site31R -0.305501  0.736754  0.449542 -0.680   0.4968    
## site32R -0.339344  0.712237  0.368543 -0.921   0.3572    
## site33R -0.323360  0.723713  0.332404 -0.973   0.3307    
## site34R -0.127196  0.880561  0.313871 -0.405   0.6853    
## site35R -0.098211  0.906458  0.333947 -0.294   0.7687    
## site36R -0.208571  0.811744  0.338918 -0.615   0.5383    
## site37R -0.162521  0.849998  0.333961 -0.487   0.6265    
## site38R -0.281466  0.754677  0.329039 -0.855   0.3923    
## site39R -0.201714  0.817329  0.360102 -0.560   0.5754    
## site40R  0.106452  1.112324  0.357942  0.297   0.7662    
## site41R -0.203641  0.815755  0.368610 -0.552   0.5806    
## site42R  0.086119  1.089936  0.359994  0.239   0.8109    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## OM         1.0272     0.9735    0.9491     1.112
## dx_yr      1.1108     0.9003    1.0816     1.141
## site02R    1.9048     0.5250    0.9853     3.682
## site03R    0.7792     1.2834    0.3946     1.539
## site04R    0.8169     1.2242    0.3384     1.972
## site05R    0.7255     1.3784    0.3512     1.499
## site06R    0.9664     1.0348    0.5104     1.830
## site07R    0.7361     1.3584    0.4049     1.338
## site08R        NA         NA        NA        NA
## site09R    0.9873     1.0129    0.5073     1.922
## site10R    0.8907     1.1227    0.4164     1.905
## site11R    0.9146     1.0934    0.5046     1.658
## site12R    0.9405     1.0633    0.4875     1.814
## site13R    0.7504     1.3327    0.4151     1.356
## site14R    0.7084     1.4116    0.2016     2.490
## site15R    1.0301     0.9708    0.5384     1.971
## site16R    1.1547     0.8660    0.5879     2.268
## site17R    1.0325     0.9685    0.5095     2.092
## site18R    0.7576     1.3199    0.4023     1.427
## site19R    1.0201     0.9803    0.5110     2.037
## site20R    0.7877     1.2695    0.3988     1.556
## site21R    0.7369     1.3570    0.3922     1.385
## site22R    0.9485     1.0543    0.5215     1.725
## site23R    0.8552     1.1693    0.4545     1.609
## site24R    0.8886     1.1254    0.4636     1.703
## site25R    0.9922     1.0078    0.5248     1.876
## site26R    0.6820     1.4662    0.3468     1.341
## site27R    1.0285     0.9723    0.2320     4.560
## site28R    0.8804     1.1359    0.3757     2.063
## site29R    0.7714     1.2964    0.3293     1.807
## site30R    0.8163     1.2250    0.4268     1.561
## site31R    0.7368     1.3573    0.3053     1.778
## site32R    0.7122     1.4040    0.3459     1.467
## site33R    0.7237     1.3818    0.3772     1.388
## site34R    0.8806     1.1356    0.4760     1.629
## site35R    0.9065     1.1032    0.4711     1.744
## site36R    0.8117     1.2319    0.4178     1.577
## site37R    0.8500     1.1765    0.4417     1.636
## site38R    0.7547     1.3251    0.3960     1.438
## site39R    0.8173     1.2235    0.4035     1.655
## site40R    1.1123     0.8990    0.5515     2.243
## site41R    0.8158     1.2259    0.3961     1.680
## site42R    1.0899     0.9175    0.5382     2.207
## 
## Concordance= 0.59  (se = 0.009 )
## Likelihood ratio test= 111  on 42 df,   p=4e-08
## Wald test            = 109.4  on 42 df,   p=7e-08
## Score (logrank) test = 113.7  on 42 df,   p=2e-08

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ OM + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=PFF_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = PFF_IPF, 
##     id = ID)
## 
##   n= 53867, number of events= 1131 
##    (849 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)      z Pr(>|z|)    
## OM                  0.0326936  1.0332339  0.0415005  0.788   0.4308    
## dx_yr               0.1040354  1.1096397  0.0140529  7.403 1.33e-13 ***
## age_dx             -0.0002621  0.9997380  0.0039353 -0.067   0.9469    
## sexM                0.0606268  1.0625024  0.0713010  0.850   0.3952    
## dich_RaceNon-White -0.0629281  0.9390110  0.1262292 -0.499   0.6181    
## smokeHxEver         0.0430889  1.0440307  0.0646643  0.666   0.5052    
## disadv             -0.0629225  0.9390163  0.1128946 -0.557   0.5773    
## site02R             0.5850645  1.7951067  0.3495436  1.674   0.0942 .  
## site03R            -0.2859287  0.7513162  0.3598005 -0.795   0.4268    
## site04R            -0.2624054  0.7691991  0.4597067 -0.571   0.5681    
## site05R            -0.3748213  0.6874121  0.3835164 -0.977   0.3284    
## site06R            -0.0758819  0.9269257  0.3416560 -0.222   0.8242    
## site07R            -0.3550241  0.7011566  0.3192538 -1.112   0.2661    
## site08R                    NA         NA  0.0000000     NA       NA    
## site09R            -0.0995772  0.9052200  0.3543448 -0.281   0.7787    
## site10R            -0.1704998  0.8432433  0.4002159 -0.426   0.6701    
## site11R            -0.1574578  0.8543128  0.3230130 -0.487   0.6259    
## site12R            -0.1066097  0.8988764  0.3481302 -0.306   0.7594    
## site13R            -0.3507132  0.7041857  0.3154205 -1.112   0.2662    
## site14R            -0.4139160  0.6610565  0.6483655 -0.638   0.5232    
## site15R            -0.0249939  0.9753159  0.3471927 -0.072   0.9426    
## site16R             0.0935601  1.0980766  0.3557467  0.263   0.7926    
## site17R            -0.0323214  0.9681953  0.3741427 -0.086   0.9312    
## site18R            -0.3174873  0.7279759  0.3388361 -0.937   0.3488    
## site19R            -0.0255824  0.9747420  0.3640588 -0.070   0.9440    
## site20R            -0.2954545  0.7441933  0.3608436 -0.819   0.4129    
## site21R            -0.3583681  0.6988158  0.3367501 -1.064   0.2872    
## site22R            -0.1118308  0.8941956  0.3192340 -0.350   0.7261    
## site23R            -0.2256642  0.7979861  0.3384212 -0.667   0.5049    
## site24R            -0.1763113  0.8383570  0.3448062 -0.511   0.6091    
## site25R            -0.0460385  0.9550052  0.3382344 -0.136   0.8917    
## site26R            -0.4170714  0.6589739  0.3574496 -1.167   0.2433    
## site27R            -0.0850962  0.9184239  0.7706431 -0.110   0.9121    
## site28R            -0.1662541  0.8468310  0.4464182 -0.372   0.7096    
## site29R            -0.3178267  0.7277289  0.4449827 -0.714   0.4751    
## site30R            -0.2629715  0.7687638  0.3439153 -0.765   0.4445    
## site31R            -0.3607265  0.6971697  0.4584399 -0.787   0.4314    
## site32R            -0.4062812  0.6661228  0.3903967 -1.041   0.2980    
## site33R            -0.3652165  0.6940464  0.3477144 -1.050   0.2936    
## site34R            -0.1737964  0.8404680  0.3286392 -0.529   0.5969    
## site35R            -0.1638864  0.8488384  0.3481140 -0.471   0.6378    
## site36R            -0.2491257  0.7794820  0.3541058 -0.704   0.4817    
## site37R            -0.2119288  0.8090223  0.3468373 -0.611   0.5412    
## site38R            -0.3240544  0.7232109  0.3427608 -0.945   0.3444    
## site39R            -0.2593504  0.7715527  0.3728587 -0.696   0.4867    
## site40R             0.0571100  1.0587722  0.3720264  0.154   0.8780    
## site41R            -0.2480017  0.7803586  0.3791269 -0.654   0.5130    
## site42R             0.0139787  1.0140769  0.3720564  0.038   0.9700    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## OM                    1.0332     0.9678    0.9525     1.121
## dx_yr                 1.1096     0.9012    1.0795     1.141
## age_dx                0.9997     1.0003    0.9921     1.007
## sexM                  1.0625     0.9412    0.9239     1.222
## dich_RaceNon-White    0.9390     1.0650    0.7332     1.203
## smokeHxEver           1.0440     0.9578    0.9198     1.185
## disadv                0.9390     1.0649    0.7526     1.172
## site02R               1.7951     0.5571    0.9048     3.561
## site03R               0.7513     1.3310    0.3712     1.521
## site04R               0.7692     1.3001    0.3124     1.894
## site05R               0.6874     1.4547    0.3242     1.458
## site06R               0.9269     1.0788    0.4745     1.811
## site07R               0.7012     1.4262    0.3750     1.311
## site08R                   NA         NA        NA        NA
## site09R               0.9052     1.1047    0.4520     1.813
## site10R               0.8432     1.1859    0.3848     1.848
## site11R               0.8543     1.1705    0.4536     1.609
## site12R               0.8989     1.1125    0.4543     1.778
## site13R               0.7042     1.4201    0.3795     1.307
## site14R               0.6611     1.5127    0.1855     2.356
## site15R               0.9753     1.0253    0.4939     1.926
## site16R               1.0981     0.9107    0.5468     2.205
## site17R               0.9682     1.0328    0.4650     2.016
## site18R               0.7280     1.3737    0.3747     1.414
## site19R               0.9747     1.0259    0.4775     1.990
## site20R               0.7442     1.3437    0.3669     1.510
## site21R               0.6988     1.4310    0.3612     1.352
## site22R               0.8942     1.1183    0.4783     1.672
## site23R               0.7980     1.2532    0.4111     1.549
## site24R               0.8384     1.1928    0.4265     1.648
## site25R               0.9550     1.0471    0.4922     1.853
## site26R               0.6590     1.5175    0.3270     1.328
## site27R               0.9184     1.0888    0.2028     4.159
## site28R               0.8468     1.1809    0.3530     2.031
## site29R               0.7277     1.3741    0.3042     1.741
## site30R               0.7688     1.3008    0.3918     1.508
## site31R               0.6972     1.4344    0.2839     1.712
## site32R               0.6661     1.5012    0.3099     1.432
## site33R               0.6940     1.4408    0.3511     1.372
## site34R               0.8405     1.1898    0.4413     1.601
## site35R               0.8488     1.1781    0.4291     1.679
## site36R               0.7795     1.2829    0.3894     1.560
## site37R               0.8090     1.2361    0.4100     1.597
## site38R               0.7232     1.3827    0.3694     1.416
## site39R               0.7716     1.2961    0.3715     1.602
## site40R               1.0588     0.9445    0.5107     2.195
## site41R               0.7804     1.2815    0.3712     1.641
## site42R               1.0141     0.9861    0.4891     2.103
## 
## Concordance= 0.591  (se = 0.009 )
## Likelihood ratio test= 111.8  on 47 df,   p=3e-07
## Wald test            = 110.6  on 47 df,   p=5e-07
## Score (logrank) test = 114.9  on 47 df,   p=1e-07

91.6.3 IPF-Only OM - CARE-PF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ OM + dx_yr + site, data=CARE_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM + dx_yr + site, 
##     data = CARE_IPF, id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef)      z Pr(>|z|)    
## OM      -0.06709   0.93511  0.03259 -2.058  0.03956 *  
## dx_yr    0.81346   2.25570  0.03229 25.191  < 2e-16 ***
## site102  0.21991   1.24596  0.14722  1.494  0.13526    
## site103  0.52353   1.68797  0.11915  4.394 1.11e-05 ***
## site104  0.38792   1.47392  0.13938  2.783  0.00538 ** 
## site105  0.09127   1.09556  0.12629  0.723  0.46988    
## site106  0.14725   1.15865  0.11987  1.228  0.21929    
## site107  0.13464   1.14413  0.19790  0.680  0.49626    
## site108 -0.42351   0.65475  0.26376 -1.606  0.10835    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## OM         0.9351     1.0694    0.8772    0.9968
## dx_yr      2.2557     0.4433    2.1174    2.4031
## site102    1.2460     0.8026    0.9337    1.6627
## site103    1.6880     0.5924    1.3364    2.1320
## site104    1.4739     0.6785    1.1216    1.9369
## site105    1.0956     0.9128    0.8553    1.4033
## site106    1.1586     0.8631    0.9160    1.4655
## site107    1.1441     0.8740    0.7763    1.6863
## site108    0.6547     1.5273    0.3904    1.0980
## 
## Concordance= 0.773  (se = 0.01 )
## Likelihood ratio test= 1056  on 9 df,   p=<2e-16
## Wald test            = 670  on 9 df,   p=<2e-16
## Score (logrank) test = 587.8  on 9 df,   p=<2e-16

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ OM + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=CARE_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = CARE_IPF, 
##     id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## OM                 -0.075853  0.926953  0.033281 -2.279  0.02266 *  
## dx_yr               0.813131  2.254956  0.032448 25.059  < 2e-16 ***
## age_dx              0.012625  1.012705  0.004511  2.799  0.00513 ** 
## sexF               -0.040532  0.960278  0.077048 -0.526  0.59884    
## dich_RaceNon-White -0.141492  0.868062  0.110343 -1.282  0.19974    
## smokeHxFormer      -0.002835  0.997169  0.081474 -0.035  0.97224    
## smokeHxAlways       0.049157  1.050385  0.168933  0.291  0.77106    
## smokeHxUnknown      0.437822  1.549328  0.734808  0.596  0.55129    
## disadv              0.021065  1.021288  0.130396  0.162  0.87166    
## site102             0.206241  1.229050  0.150076  1.374  0.16937    
## site103             0.538568  1.713550  0.121814  4.421 9.81e-06 ***
## site104             0.397174  1.487614  0.139924  2.838  0.00453 ** 
## site105             0.113843  1.120576  0.129412  0.880  0.37903    
## site106             0.119709  1.127169  0.121016  0.989  0.32256    
## site107             0.072980  1.075709  0.204978  0.356  0.72181    
## site108            -0.426751  0.652626  0.267118 -1.598  0.11013    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## OM                    0.9270     1.0788    0.8684    0.9894
## dx_yr                 2.2550     0.4435    2.1160    2.4030
## age_dx                1.0127     0.9875    1.0038    1.0217
## sexF                  0.9603     1.0414    0.8257    1.1168
## dich_RaceNon-White    0.8681     1.1520    0.6992    1.0776
## smokeHxFormer         0.9972     1.0028    0.8500    1.1698
## smokeHxAlways         1.0504     0.9520    0.7543    1.4627
## smokeHxUnknown        1.5493     0.6454    0.3670    6.5405
## disadv                1.0213     0.9792    0.7910    1.3187
## site102               1.2290     0.8136    0.9158    1.6494
## site103               1.7136     0.5836    1.3496    2.1756
## site104               1.4876     0.6722    1.1308    1.9570
## site105               1.1206     0.8924    0.8695    1.4441
## site106               1.1272     0.8872    0.8892    1.4289
## site107               1.0757     0.9296    0.7198    1.6076
## site108               0.6526     1.5323    0.3866    1.1016
## 
## Concordance= 0.775  (se = 0.01 )
## Likelihood ratio test= 1069  on 16 df,   p=<2e-16
## Wald test            = 677.9  on 16 df,   p=<2e-16
## Score (logrank) test = 594.1  on 16 df,   p=<2e-16

91.6.4 IPF-Only OM - Combined Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ OM + dx_yr + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM + dx_yr + site, 
##     data = All_IPF, id = ID, cluster = cohort)
## 
##   n= 123367, number of events= 2749 
##    (723 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## OM       0.008346  1.008381  0.021735  0.047598   0.175 0.860813    
## dx_yr    0.104256  1.109884  0.006955  0.074653   1.397 0.162550    
## site02R  0.664066  1.942675  0.334586  0.051497  12.895  < 2e-16 ***
## site03R -0.275403  0.759266  0.347078  0.019242 -14.313  < 2e-16 ***
## site04R -0.224335  0.799047  0.449438  0.091090  -2.463 0.013786 *  
## site05R -0.335114  0.715257  0.369060  0.036067  -9.291  < 2e-16 ***
## site06R -0.049078  0.952106  0.325096  0.094334  -0.520 0.602880    
## site07R -0.324164  0.723131  0.304795  0.094544  -3.429 0.000606 ***
## site09R -0.035434  0.965186  0.339723  0.017681  -2.004 0.045061 *  
## site1    0.242521  1.274458  0.281363  0.119281   2.033 0.042034 *  
## site101 -0.109665  0.896134  0.293302  0.136328  -0.804 0.421154    
## site102 -0.236867  0.789096  0.297880  0.110874  -2.136 0.032650 *  
## site103  0.097367  1.102265  0.286293  0.088352   1.102 0.270443    
## site104 -0.190366  0.826656  0.294726  0.079796  -2.386 0.017049 *  
## site105  0.080056  1.083348  0.289903  0.120477   0.664 0.506375    
## site106 -0.212939  0.808206  0.286537  0.073151  -2.911 0.003604 ** 
## site107  0.211391  1.235396  0.326950  0.138460   1.527 0.126828    
## site108  0.152312  1.164524  0.369607  0.222935   0.683 0.494473    
## site10R -0.141991  0.867629  0.385921  0.120967  -1.174 0.240478    
## site11R -0.069778  0.932601  0.300606  0.006433 -10.848  < 2e-16 ***
## site12R -0.095622  0.908807  0.334148  0.021122  -4.527 5.98e-06 ***
## site13R -0.298481  0.741945  0.301811  0.094185  -3.169 0.001529 ** 
## site14R -0.365042  0.694167  0.640720  0.111768  -3.266 0.001091 ** 
## site15R  0.030659  1.031134  0.330657  0.014133   2.169 0.030060 *  
## site16R  0.140725  1.151108  0.344387  0.031560   4.459 8.24e-06 ***
## site17R  0.038494  1.039245  0.360084  0.014346   2.683 0.007292 ** 
## site18R -0.264889  0.767291  0.321827  0.133761  -1.980 0.047668 *  
## site19R -0.022981  0.977281  0.350635  0.057136  -0.402 0.687526    
## site20R -0.272055  0.761813  0.347097  0.023017 -11.820  < 2e-16 ***
## site21R -0.320116  0.726065  0.320161  0.015096 -21.205  < 2e-16 ***
## site22R -0.070026  0.932369  0.305090  0.012464  -5.618 1.93e-08 ***
## site23R -0.165905  0.847126  0.322482  0.008582 -19.333  < 2e-16 ***
## site24R -0.160898  0.851379  0.330845  0.027024  -5.954 2.62e-09 ***
## site25R -0.023385  0.976886  0.324883  0.041771  -0.560 0.575587    
## site26R -0.410523  0.663303  0.344560  0.099356  -4.132 3.60e-05 ***
## site27R -0.014297  0.985805  0.759679  0.120008  -0.119 0.905169    
## site28R -0.146860  0.863415  0.433952  0.037145  -3.954 7.69e-05 ***
## site29R -0.301680  0.739574  0.433826  0.069702  -4.328 1.50e-05 ***
## site30R -0.213175  0.808014  0.330584  0.093118  -2.289 0.022062 *  
## site31R -0.334912  0.715401  0.449440  0.075244  -4.451 8.55e-06 ***
## site32R -0.352451  0.702963  0.368485  0.037966  -9.283  < 2e-16 ***
## site33R -0.350559  0.704294  0.332178  0.095597  -3.667 0.000245 ***
## site34R -0.157525  0.854256  0.313516  0.059478  -2.648 0.008086 ** 
## site35R -0.114006  0.892253  0.333822  0.046095  -2.473 0.013387 *  
## site36R -0.226130  0.797615  0.338070  0.032639  -6.928 4.26e-12 ***
## site37R -0.185355  0.830809  0.333883  0.038438  -4.822 1.42e-06 ***
## site38R -0.298606  0.741852  0.328981  0.024737 -12.071  < 2e-16 ***
## site39R -0.220709  0.801950  0.360016  0.013721 -16.085  < 2e-16 ***
## site40R  0.148627  1.160240  0.346586  0.017525   8.481  < 2e-16 ***
## site41R -0.214543  0.806910  0.368524  0.028918  -7.419 1.18e-13 ***
## site42R  0.082961  1.086499  0.359962  0.023821   3.483 0.000496 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## OM         1.0084     0.9917    0.9186    1.1070
## dx_yr      1.1099     0.9010    0.9588    1.2848
## site02R    1.9427     0.5148    1.7562    2.1490
## site03R    0.7593     1.3171    0.7312    0.7884
## site04R    0.7990     1.2515    0.6684    0.9552
## site05R    0.7153     1.3981    0.6664    0.7676
## site06R    0.9521     1.0503    0.7914    1.1455
## site07R    0.7231     1.3829    0.6008    0.8703
## site09R    0.9652     1.0361    0.9323    0.9992
## site1      1.2745     0.7846    1.0088    1.6101
## site101    0.8961     1.1159    0.6860    1.1706
## site102    0.7891     1.2673    0.6350    0.9806
## site103    1.1023     0.9072    0.9270    1.3107
## site104    0.8267     1.2097    0.7070    0.9666
## site105    1.0833     0.9231    0.8555    1.3719
## site106    0.8082     1.2373    0.7003    0.9328
## site107    1.2354     0.8095    0.9418    1.6206
## site108    1.1645     0.8587    0.7523    1.8026
## site10R    0.8676     1.1526    0.6845    1.0998
## site11R    0.9326     1.0723    0.9209    0.9444
## site12R    0.9088     1.1003    0.8720    0.9472
## site13R    0.7419     1.3478    0.6169    0.8924
## site14R    0.6942     1.4406    0.5576    0.8642
## site15R    1.0311     0.9698    1.0030    1.0601
## site16R    1.1511     0.8687    1.0821    1.2246
## site17R    1.0392     0.9622    1.0104    1.0689
## site18R    0.7673     1.3033    0.5903    0.9973
## site19R    0.9773     1.0232    0.8737    1.0931
## site20R    0.7618     1.3127    0.7282    0.7970
## site21R    0.7261     1.3773    0.7049    0.7479
## site22R    0.9324     1.0725    0.9099    0.9554
## site23R    0.8471     1.1805    0.8330    0.8615
## site24R    0.8514     1.1746    0.8075    0.8977
## site25R    0.9769     1.0237    0.9001    1.0602
## site26R    0.6633     1.5076    0.5459    0.8059
## site27R    0.9858     1.0144    0.7792    1.2472
## site28R    0.8634     1.1582    0.8028    0.9286
## site29R    0.7396     1.3521    0.6451    0.8478
## site30R    0.8080     1.2376    0.6732    0.9698
## site31R    0.7154     1.3978    0.6173    0.8291
## site32R    0.7030     1.4226    0.6526    0.7573
## site33R    0.7043     1.4199    0.5840    0.8494
## site34R    0.8543     1.1706    0.7603    0.9599
## site35R    0.8923     1.1208    0.8152    0.9766
## site36R    0.7976     1.2537    0.7482    0.8503
## site37R    0.8308     1.2036    0.7705    0.8958
## site38R    0.7419     1.3480    0.7067    0.7787
## site39R    0.8019     1.2470    0.7807    0.8238
## site40R    1.1602     0.8619    1.1211    1.2008
## site41R    0.8069     1.2393    0.7624    0.8540
## site42R    1.0865     0.9204    1.0369    1.1384
## 
## Concordance= 0.595  (se = 0.043 )
## Likelihood ratio test= 347.9  on 51 df,   p=<2e-16
## Wald test            = 148.6  on 51 df,   p=2e-11
## Score (logrank) test = 337.6  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ OM + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ OM + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = All_IPF, 
##     id = ID, cluster = cohort)
## 
##   n= 121241, number of events= 2709 
##    (2849 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## OM                  0.004681  1.004692  0.022151  0.041890   0.112 0.911025    
## dx_yr               0.100974  1.106248  0.007326  0.079954   1.263 0.206620    
## age_dx              0.004216  1.004225  0.002447  0.002208   1.910 0.056185 .  
## sexF               -0.158003  0.853847  0.044865  0.084182  -1.877 0.060527 .  
## dich_RaceNon-White  0.017006  1.017151  0.068701  0.058620   0.290 0.771734    
## smokeHxFormer       0.099364  1.104468  0.059381  0.041267   2.408 0.016048 *  
## smokeHxAlways      -0.031110  0.969369  0.138358  0.117532  -0.265 0.791247    
## smokeHxUnknown      0.316941  1.372921  0.153843  0.222183   1.426 0.153728    
## smokeHxEver         0.036076  1.036735  0.064257  0.008292   4.351 1.36e-05 ***
## disadv              0.041195  1.042056  0.070848  0.117076   0.352 0.724936    
## site02R             0.652366  1.920078  0.345392  0.067782   9.625  < 2e-16 ***
## site03R            -0.253805  0.775843  0.357290  0.017898 -14.180  < 2e-16 ***
## site04R            -0.219520  0.802904  0.457637  0.102680  -2.138 0.032525 *  
## site05R            -0.328336  0.720121  0.379714  0.034133  -9.619  < 2e-16 ***
## site06R            -0.059713  0.942035  0.337836  0.121345  -0.492 0.622655    
## site07R            -0.331678  0.717718  0.316470  0.088483  -3.748 0.000178 ***
## site09R            -0.083470  0.919919  0.352062  0.001438 -58.031  < 2e-16 ***
## site1               0.175666  1.192040  0.297702  0.164465   1.068 0.285473    
## site101            -0.161604  0.850778  0.308371  0.115385  -1.401 0.161344    
## site102            -0.307594  0.735214  0.313627  0.096765  -3.179 0.001479 ** 
## site103             0.020282  1.020489  0.301990  0.045563   0.445 0.656222    
## site104            -0.255829  0.774275  0.310609  0.045097  -5.673 1.40e-08 ***
## site105             0.022701  1.022961  0.307198  0.101970   0.223 0.823825    
## site106            -0.290768  0.747689  0.302663  0.049020  -5.932 3.00e-09 ***
## site107             0.118892  1.126248  0.342631  0.110474   1.076 0.281837    
## site108             0.075676  1.078613  0.381971  0.192640   0.393 0.694439    
## site10R            -0.211530  0.809345  0.396500  0.087528  -2.417 0.015661 *  
## site11R            -0.099368  0.905409  0.315281  0.018643  -5.330 9.82e-08 ***
## site12R            -0.118999  0.887809  0.345131  0.011043 -10.776  < 2e-16 ***
## site13R            -0.336484  0.714278  0.313407  0.088850  -3.787 0.000152 ***
## site14R            -0.372734  0.688848  0.646409  0.094223  -3.956 7.63e-05 ***
## site15R            -0.008039  0.991994  0.342920  0.036392  -0.221 0.825181    
## site16R             0.143184  1.153943  0.354452  0.047487   3.015 0.002568 ** 
## site17R             0.025963  1.026303  0.370946  0.027066   0.959 0.337428    
## site18R            -0.294331  0.745030  0.333908  0.108701  -2.708 0.006775 ** 
## site19R            -0.069435  0.932921  0.361039  0.036169  -1.920 0.054889 .  
## site20R            -0.269864  0.763483  0.358311  0.012276 -21.982  < 2e-16 ***
## site21R            -0.354576  0.701471  0.331817  0.019419 -18.259  < 2e-16 ***
## site22R            -0.081510  0.921723  0.316971  0.013788  -5.912 3.38e-09 ***
## site23R            -0.223322  0.799857  0.336226  0.022857  -9.770  < 2e-16 ***
## site24R            -0.175609  0.838946  0.342274  0.030820  -5.698 1.21e-08 ***
## site25R            -0.039470  0.961299  0.336777  0.052945  -0.745 0.455975    
## site26R            -0.451676  0.636560  0.355351  0.067788  -6.663 2.68e-11 ***
## site27R             0.000429  1.000429  0.766948  0.123980   0.003 0.997239    
## site28R            -0.144701  0.865281  0.442781  0.042004  -3.445 0.000571 ***
## site29R            -0.308749  0.734365  0.442857  0.091163  -3.387 0.000707 ***
## site30R            -0.276432  0.758485  0.342778  0.080637  -3.428 0.000608 ***
## site31R            -0.371206  0.689902  0.457592  0.059850  -6.202 5.57e-10 ***
## site32R            -0.395238  0.673520  0.388759  0.020311 -19.460  < 2e-16 ***
## site33R            -0.391568  0.675996  0.345261  0.066884  -5.854 4.79e-09 ***
## site34R            -0.173880  0.840397  0.326085  0.051279  -3.391 0.000697 ***
## site35R            -0.166380  0.846724  0.346841  0.025383  -6.555 5.57e-11 ***
## site36R            -0.229358  0.795044  0.349409  0.035515  -6.458 1.06e-10 ***
## site37R            -0.201270  0.817692  0.344772  0.038922  -5.171 2.33e-07 ***
## site38R            -0.304846  0.737237  0.340284  0.003326 -91.652  < 2e-16 ***
## site39R            -0.262556  0.769083  0.370552  0.016135 -16.272  < 2e-16 ***
## site40R             0.114435  1.121240  0.357337  0.031877   3.590 0.000331 ***
## site41R            -0.253223  0.776295  0.377847  0.016139 -15.690  < 2e-16 ***
## site42R             0.057802  1.059505  0.370059  0.027661   2.090 0.036650 *  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## OM                    1.0047     0.9953    0.9255    1.0907
## dx_yr                 1.1062     0.9040    0.9458    1.2939
## age_dx                1.0042     0.9958    0.9999    1.0086
## sexF                  0.8538     1.1712    0.7240    1.0070
## dich_RaceNon-White    1.0172     0.9831    0.9068    1.1410
## smokeHxFormer         1.1045     0.9054    1.0187    1.1975
## smokeHxAlways         0.9694     1.0316    0.7699    1.2205
## smokeHxUnknown        1.3729     0.7284    0.8882    2.1221
## smokeHxEver           1.0367     0.9646    1.0200    1.0537
## disadv                1.0421     0.9596    0.8284    1.3108
## site02R               1.9201     0.5208    1.6812    2.1929
## site03R               0.7758     1.2889    0.7491    0.8035
## site04R               0.8029     1.2455    0.6565    0.9819
## site05R               0.7201     1.3887    0.6735    0.7699
## site06R               0.9420     1.0615    0.7426    1.1950
## site07R               0.7177     1.3933    0.6034    0.8536
## site09R               0.9199     1.0871    0.9173    0.9225
## site1                 1.1920     0.8389    0.8636    1.6454
## site101               0.8508     1.1754    0.6786    1.0667
## site102               0.7352     1.3601    0.6082    0.8888
## site103               1.0205     0.9799    0.9333    1.1158
## site104               0.7743     1.2915    0.7088    0.8458
## site105               1.0230     0.9776    0.8376    1.2493
## site106               0.7477     1.3375    0.6792    0.8231
## site107               1.1262     0.8879    0.9070    1.3985
## site108               1.0786     0.9271    0.7394    1.5734
## site10R               0.8093     1.2356    0.6818    0.9608
## site11R               0.9054     1.1045    0.8729    0.9391
## site12R               0.8878     1.1264    0.8688    0.9072
## site13R               0.7143     1.4000    0.6001    0.8502
## site14R               0.6888     1.4517    0.5727    0.8286
## site15R               0.9920     1.0081    0.9237    1.0653
## site16R               1.1539     0.8666    1.0514    1.2665
## site17R               1.0263     0.9744    0.9733    1.0822
## site18R               0.7450     1.3422    0.6021    0.9219
## site19R               0.9329     1.0719    0.8691    1.0015
## site20R               0.7635     1.3098    0.7453    0.7821
## site21R               0.7015     1.4256    0.6753    0.7287
## site22R               0.9217     1.0849    0.8971    0.9470
## site23R               0.7999     1.2502    0.7648    0.8365
## site24R               0.8389     1.1920    0.7898    0.8912
## site25R               0.9613     1.0403    0.8665    1.0664
## site26R               0.6366     1.5709    0.5574    0.7270
## site27R               1.0004     0.9996    0.7846    1.2756
## site28R               0.8653     1.1557    0.7969    0.9395
## site29R               0.7344     1.3617    0.6142    0.8780
## site30R               0.7585     1.3184    0.6476    0.8884
## site31R               0.6899     1.4495    0.6135    0.7758
## site32R               0.6735     1.4847    0.6472    0.7009
## site33R               0.6760     1.4793    0.5929    0.7707
## site34R               0.8404     1.1899    0.7600    0.9293
## site35R               0.8467     1.1810    0.8056    0.8899
## site36R               0.7950     1.2578    0.7416    0.8524
## site37R               0.8177     1.2230    0.7576    0.8825
## site38R               0.7372     1.3564    0.7324    0.7421
## site39R               0.7691     1.3002    0.7451    0.7938
## site40R               1.1212     0.8919    1.0533    1.1935
## site41R               0.7763     1.2882    0.7521    0.8012
## site42R               1.0595     0.9438    1.0036    1.1185
## 
## Concordance= 0.598  (se = 0.039 )
## Likelihood ratio test= 379.9  on 59 df,   p=<2e-16
## Wald test            = 35.12  on 59 df,   p=1
## Score (logrank) test = 374.4  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

91.7 IPF-Only SS

91.7.1 IPF-Only SS - Simmons Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ SS + dx_yr, data=Simm_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS + dx_yr, data = Simm_IPF, 
##     id = ID)
## 
##   n= 31982, number of events= 695 
##    (446 observations deleted due to missingness)
## 
##          coef exp(coef) se(coef)     z Pr(>|z|)   
## SS    0.47439   1.60703  0.27965 1.696   0.0898 . 
## dx_yr 0.02377   1.02406  0.00845 2.813   0.0049 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##       exp(coef) exp(-coef) lower .95 upper .95
## SS        1.607     0.6223    0.9289     2.780
## dx_yr     1.024     0.9765    1.0072     1.041
## 
## Concordance= 0.524  (se = 0.012 )
## Likelihood ratio test= 10.26  on 2 df,   p=0.006
## Wald test            = 10.68  on 2 df,   p=0.005
## Score (logrank) test = 10.71  on 2 df,   p=0.005

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SS + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv, data=Simm_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv, data = Simm_IPF, id = ID)
## 
##   n= 30488, number of events= 670 
##    (1940 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## SS                  0.368463  1.445511  0.300869  1.225  0.22070    
## dx_yr               0.011621  1.011689  0.009072  1.281  0.20020    
## age_dx              0.002711  1.002715  0.004576  0.593  0.55346    
## sexF               -0.372239  0.689189  0.089029 -4.181  2.9e-05 ***
## dich_RaceNon-White  0.155189  1.167878  0.128361  1.209  0.22666    
## smokeHxFormer       0.084484  1.088156  0.091868  0.920  0.35777    
## smokeHxAlways      -0.408091  0.664918  0.266268 -1.533  0.12537    
## smokeHxUnknown      0.495109  1.640676  0.170364  2.906  0.00366 ** 
## disadv              0.382720  1.466267  0.133294  2.871  0.00409 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SS                    1.4455     0.6918    0.8015    2.6069
## dx_yr                 1.0117     0.9884    0.9939    1.0298
## age_dx                1.0027     0.9973    0.9938    1.0117
## sexF                  0.6892     1.4510    0.5788    0.8206
## dich_RaceNon-White    1.1679     0.8563    0.9081    1.5020
## smokeHxFormer         1.0882     0.9190    0.9089    1.3028
## smokeHxAlways         0.6649     1.5039    0.3946    1.1205
## smokeHxUnknown        1.6407     0.6095    1.1749    2.2911
## disadv                1.4663     0.6820    1.1292    1.9040
## 
## Concordance= 0.592  (se = 0.012 )
## Likelihood ratio test= 53.04  on 9 df,   p=3e-08
## Wald test            = 54.09  on 9 df,   p=2e-08
## Score (logrank) test = 54.86  on 9 df,   p=1e-08

91.7.2 IPF-Only SS - PFF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ SS + dx_yr + site, data=PFF_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS + dx_yr + site, 
##     data = PFF_IPF, id = ID)
## 
##   n= 54499, number of events= 1146 
##    (217 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef)      z Pr(>|z|)    
## SS       0.012132  1.012206  0.146170  0.083    0.934    
## dx_yr    0.104169  1.109788  0.013561  7.682 1.57e-14 ***
## site02R  0.643993  1.904068  0.381077  1.690    0.091 .  
## site03R -0.243484  0.783892  0.347691 -0.700    0.484    
## site04R -0.205944  0.813879  0.449637 -0.458    0.647    
## site05R -0.296800  0.743193  0.369167 -0.804    0.421    
## site06R -0.020232  0.979971  0.325778 -0.062    0.950    
## site07R -0.301975  0.739356  0.305022 -0.990    0.322    
## site08R        NA        NA  0.000000     NA       NA    
## site09R -0.013630  0.986463  0.340068 -0.040    0.968    
## site10R -0.155023  0.856395  0.402947 -0.385    0.700    
## site11R -0.056393  0.945168  0.299531 -0.188    0.851    
## site12R -0.077962  0.924999  0.334777 -0.233    0.816    
## site13R -0.284715  0.752229  0.302073 -0.943    0.346    
## site14R -0.349780  0.704843  0.641269 -0.545    0.585    
## site15R  0.041796  1.042682  0.331362  0.126    0.900    
## site16R  0.141405  1.151891  0.345124  0.410    0.682    
## site17R  0.044760  1.045777  0.360440  0.124    0.901    
## site18R -0.266729  0.765881  0.323816 -0.824    0.410    
## site19R -0.008356  0.991679  0.350097 -0.024    0.981    
## site20R -0.240895  0.785924  0.347874 -0.692    0.489    
## site21R -0.278103  0.757219  0.319515 -0.870    0.384    
## site22R -0.054596  0.946867  0.305497 -0.179    0.858    
## site23R -0.151175  0.859697  0.322467 -0.469    0.639    
## site24R -0.132746  0.875688  0.331557 -0.400    0.689    
## site25R -0.017925  0.982235  0.363597 -0.049    0.961    
## site26R -0.400218  0.670174  0.354289 -1.130    0.259    
## site27R  0.023814  1.024100  0.759872  0.031    0.975    
## site28R -0.108669  0.897027  0.434013 -0.250    0.802    
## site29R -0.277885  0.757384  0.436755 -0.636    0.525    
## site30R -0.197830  0.820509  0.330810 -0.598    0.550    
## site31R -0.307811  0.735054  0.449879 -0.684    0.494    
## site32R -0.341448  0.710740  0.368528 -0.927    0.354    
## site33R -0.332222  0.717328  0.334975 -0.992    0.321    
## site34R -0.136132  0.872727  0.313602 -0.434    0.664    
## site35R -0.104672  0.900620  0.334014 -0.313    0.754    
## site36R -0.188279  0.828384  0.337914 -0.557    0.577    
## site37R -0.157661  0.854139  0.333958 -0.472    0.637    
## site38R -0.276427  0.758489  0.328990 -0.840    0.401    
## site39R -0.196625  0.821499  0.360829 -0.545    0.586    
## site40R  0.170358  1.185730  0.346234  0.492    0.623    
## site41R -0.199369  0.819248  0.369200 -0.540    0.589    
## site42R  0.086184  1.090007  0.360004  0.239    0.811    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## SS         1.0122     0.9879    0.7601     1.348
## dx_yr      1.1098     0.9011    1.0807     1.140
## site02R    1.9041     0.5252    0.9022     4.018
## site03R    0.7839     1.2757    0.3966     1.550
## site04R    0.8139     1.2287    0.3372     1.965
## site05R    0.7432     1.3455    0.3605     1.532
## site06R    0.9800     1.0204    0.5175     1.856
## site07R    0.7394     1.3525    0.4066     1.344
## site08R        NA         NA        NA        NA
## site09R    0.9865     1.0137    0.5065     1.921
## site10R    0.8564     1.1677    0.3888     1.887
## site11R    0.9452     1.0580    0.5255     1.700
## site12R    0.9250     1.0811    0.4799     1.783
## site13R    0.7522     1.3294    0.4161     1.360
## site14R    0.7048     1.4188    0.2006     2.477
## site15R    1.0427     0.9591    0.5446     1.996
## site16R    1.1519     0.8681    0.5857     2.266
## site17R    1.0458     0.9562    0.5160     2.120
## site18R    0.7659     1.3057    0.4060     1.445
## site19R    0.9917     1.0084    0.4993     1.970
## site20R    0.7859     1.2724    0.3974     1.554
## site21R    0.7572     1.3206    0.4048     1.416
## site22R    0.9469     1.0561    0.5203     1.723
## site23R    0.8597     1.1632    0.4569     1.617
## site24R    0.8757     1.1420    0.4572     1.677
## site25R    0.9822     1.0181    0.4816     2.003
## site26R    0.6702     1.4921    0.3347     1.342
## site27R    1.0241     0.9765    0.2310     4.541
## site28R    0.8970     1.1148    0.3832     2.100
## site29R    0.7574     1.3203    0.3218     1.783
## site30R    0.8205     1.2188    0.4290     1.569
## site31R    0.7351     1.3604    0.3044     1.775
## site32R    0.7107     1.4070    0.3452     1.464
## site33R    0.7173     1.3941    0.3720     1.383
## site34R    0.8727     1.1458    0.4720     1.614
## site35R    0.9006     1.1103    0.4680     1.733
## site36R    0.8284     1.2072    0.4272     1.606
## site37R    0.8541     1.1708    0.4439     1.644
## site38R    0.7585     1.3184    0.3980     1.445
## site39R    0.8215     1.2173    0.4050     1.666
## site40R    1.1857     0.8434    0.6015     2.337
## site41R    0.8192     1.2206    0.3973     1.689
## site42R    1.0900     0.9174    0.5383     2.207
## 
## Concordance= 0.59  (se = 0.009 )
## Likelihood ratio test= 110.5  on 42 df,   p=4e-08
## Wald test            = 108.8  on 42 df,   p=8e-08
## Score (logrank) test = 113.1  on 42 df,   p=2e-08

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SS + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=PFF_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = PFF_IPF, 
##     id = ID)
## 
##   n= 53867, number of events= 1131 
##    (849 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)      z Pr(>|z|)    
## SS                  0.0218929  1.0221343  0.1491498  0.147    0.883    
## dx_yr               0.1027901  1.1082588  0.0139995  7.342  2.1e-13 ***
## age_dx             -0.0002854  0.9997146  0.0039360 -0.073    0.942    
## sexM                0.0622603  1.0642394  0.0713709  0.872    0.383    
## dich_RaceNon-White -0.0590671  0.9426435  0.1262727 -0.468    0.640    
## smokeHxEver         0.0433413  1.0442943  0.0646418  0.670    0.503    
## disadv             -0.0514455  0.9498554  0.1122989 -0.458    0.647    
## site02R             0.5864974  1.7976808  0.3948631  1.485    0.137    
## site03R            -0.2677114  0.7651286  0.3594977 -0.745    0.456    
## site04R            -0.2550642  0.7748667  0.4595574 -0.555    0.579    
## site05R            -0.3351150  0.7152558  0.3808214 -0.880    0.379    
## site06R            -0.0512778  0.9500147  0.3406824 -0.151    0.880    
## site07R            -0.3400899  0.7117064  0.3186097 -1.067    0.286    
## site08R                    NA         NA  0.0000000     NA       NA    
## site09R            -0.0909110  0.9130990  0.3544148 -0.257    0.798    
## site10R            -0.2164570  0.8053672  0.4178697 -0.518    0.604    
## site11R            -0.1087201  0.8969815  0.3169932 -0.343    0.732    
## site12R            -0.1167564  0.8898019  0.3480799 -0.335    0.737    
## site13R            -0.3370267  0.7138898  0.3148859 -1.070    0.284    
## site14R            -0.4067233  0.6658284  0.6482548 -0.627    0.530    
## site15R            -0.0040338  0.9959743  0.3470109 -0.012    0.991    
## site16R             0.0993052  1.1044033  0.3564090  0.279    0.781    
## site17R            -0.0064610  0.9935599  0.3730889 -0.017    0.986    
## site18R            -0.2986036  0.7418534  0.3393004 -0.880    0.379    
## site19R            -0.0508260  0.9504440  0.3624229 -0.140    0.888    
## site20R            -0.2868189  0.7506476  0.3610845 -0.794    0.427    
## site21R            -0.3162291  0.7288925  0.3326060 -0.951    0.342    
## site22R            -0.1041534  0.9010871  0.3193933 -0.326    0.744    
## site23R            -0.2106743  0.8100379  0.3377794 -0.624    0.533    
## site24R            -0.1831265  0.8326629  0.3447836 -0.531    0.595    
## site25R            -0.0558576  0.9456738  0.3753215 -0.149    0.882    
## site26R            -0.4353478  0.6470396  0.3680660 -1.183    0.237    
## site27R            -0.0767666  0.9261059  0.7705123 -0.100    0.921    
## site28R            -0.1339878  0.8746007  0.4447679 -0.301    0.763    
## site29R            -0.3304603  0.7185929  0.4481229 -0.737    0.461    
## site30R            -0.2477709  0.7805388  0.3433925 -0.722    0.471    
## site31R            -0.3529850  0.7025878  0.4586174 -0.770    0.441    
## site32R            -0.3959510  0.6730397  0.3901100 -1.015    0.310    
## site33R            -0.3721915  0.6892222  0.3509372 -1.061    0.289    
## site34R            -0.1744177  0.8399460  0.3285173 -0.531    0.595    
## site35R            -0.1617393  0.8506629  0.3483175 -0.464    0.642    
## site36R            -0.2150827  0.8064747  0.3516336 -0.612    0.541    
## site37R            -0.1961081  0.8219234  0.3462448 -0.566    0.571    
## site38R            -0.3096865  0.7336769  0.3422130 -0.905    0.365    
## site39R            -0.2423828  0.7847557  0.3727553 -0.650    0.516    
## site40R             0.1402910  1.1506086  0.3571384  0.393    0.694    
## site41R            -0.2351186  0.7904771  0.3794477 -0.620    0.535    
## site42R             0.0249730  1.0252875  0.3716900  0.067    0.946    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SS                    1.0221     0.9783    0.7630     1.369
## dx_yr                 1.1083     0.9023    1.0783     1.139
## age_dx                0.9997     1.0003    0.9920     1.007
## sexM                  1.0642     0.9396    0.9253     1.224
## dich_RaceNon-White    0.9426     1.0608    0.7360     1.207
## smokeHxEver           1.0443     0.9576    0.9200     1.185
## disadv                0.9499     1.0528    0.7622     1.184
## site02R               1.7977     0.5563    0.8291     3.898
## site03R               0.7651     1.3070    0.3782     1.548
## site04R               0.7749     1.2905    0.3148     1.907
## site05R               0.7153     1.3981    0.3391     1.509
## site06R               0.9500     1.0526    0.4872     1.852
## site07R               0.7117     1.4051    0.3812     1.329
## site08R                   NA         NA        NA        NA
## site09R               0.9131     1.0952    0.4559     1.829
## site10R               0.8054     1.2417    0.3551     1.827
## site11R               0.8970     1.1149    0.4819     1.670
## site12R               0.8898     1.1238    0.4498     1.760
## site13R               0.7139     1.4008    0.3851     1.323
## site14R               0.6658     1.5019    0.1869     2.372
## site15R               0.9960     1.0040    0.5045     1.966
## site16R               1.1044     0.9055    0.5492     2.221
## site17R               0.9936     1.0065    0.4782     2.064
## site18R               0.7419     1.3480    0.3815     1.443
## site19R               0.9504     1.0521    0.4671     1.934
## site20R               0.7506     1.3322    0.3699     1.523
## site21R               0.7289     1.3719    0.3798     1.399
## site22R               0.9011     1.1098    0.4818     1.685
## site23R               0.8100     1.2345    0.4178     1.570
## site24R               0.8327     1.2010    0.4236     1.637
## site25R               0.9457     1.0574    0.4532     1.973
## site26R               0.6470     1.5455    0.3145     1.331
## site27R               0.9261     1.0798    0.2045     4.193
## site28R               0.8746     1.1434    0.3658     2.091
## site29R               0.7186     1.3916    0.2986     1.730
## site30R               0.7805     1.2812    0.3982     1.530
## site31R               0.7026     1.4233    0.2860     1.726
## site32R               0.6730     1.4858    0.3133     1.446
## site33R               0.6892     1.4509    0.3465     1.371
## site34R               0.8399     1.1906    0.4412     1.599
## site35R               0.8507     1.1756    0.4298     1.684
## site36R               0.8065     1.2400    0.4048     1.607
## site37R               0.8219     1.2167    0.4170     1.620
## site38R               0.7337     1.3630    0.3752     1.435
## site39R               0.7848     1.2743    0.3780     1.629
## site40R               1.1506     0.8691    0.5714     2.317
## site41R               0.7905     1.2651    0.3758     1.663
## site42R               1.0253     0.9753    0.4948     2.124
## 
## Concordance= 0.591  (se = 0.009 )
## Likelihood ratio test= 111.2  on 47 df,   p=4e-07
## Wald test            = 109.8  on 47 df,   p=6e-07
## Score (logrank) test = 114.2  on 47 df,   p=2e-07

91.7.3 IPF-Only SS - CARE-PF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ SS + dx_yr + site, data=CARE_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS + dx_yr + site, 
##     data = CARE_IPF, id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef)      z Pr(>|z|)    
## SS      -0.12109   0.88596  0.49795 -0.243  0.80787    
## dx_yr    0.81424   2.25746  0.03235 25.169  < 2e-16 ***
## site102  0.19860   1.21970  0.17179  1.156  0.24764    
## site103  0.52716   1.69412  0.12162  4.334 1.46e-05 ***
## site104  0.37377   1.45320  0.13921  2.685  0.00725 ** 
## site105  0.06562   1.06782  0.12613  0.520  0.60289    
## site106  0.12745   1.13593  0.12023  1.060  0.28912    
## site107  0.10624   1.11209  0.19747  0.538  0.59058    
## site108 -0.41809   0.65831  0.27812 -1.503  0.13278    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## SS         0.8860     1.1287    0.3339     2.351
## dx_yr      2.2575     0.4430    2.1188     2.405
## site102    1.2197     0.8199    0.8710     1.708
## site103    1.6941     0.5903    1.3348     2.150
## site104    1.4532     0.6881    1.1062     1.909
## site105    1.0678     0.9365    0.8339     1.367
## site106    1.1359     0.8803    0.8974     1.438
## site107    1.1121     0.8992    0.7552     1.638
## site108    0.6583     1.5191    0.3817     1.135
## 
## Concordance= 0.774  (se = 0.01 )
## Likelihood ratio test= 1052  on 9 df,   p=<2e-16
## Wald test            = 665.9  on 9 df,   p=<2e-16
## Score (logrank) test = 588.3  on 9 df,   p=<2e-16

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SS + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=CARE_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = CARE_IPF, 
##     id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## SS                 -0.156608  0.855039  0.507237 -0.309  0.75751    
## dx_yr               0.813552  2.255907  0.032493 25.038  < 2e-16 ***
## age_dx              0.012007  1.012079  0.004498  2.669  0.00760 ** 
## sexF               -0.046391  0.954669  0.077129 -0.601  0.54753    
## dich_RaceNon-White -0.139844  0.869494  0.110260 -1.268  0.20469    
## smokeHxFormer       0.008827  1.008866  0.081359  0.108  0.91360    
## smokeHxAlways       0.053362  1.054811  0.168742  0.316  0.75183    
## smokeHxUnknown      0.401423  1.493949  0.734621  0.546  0.58477    
## disadv             -0.018446  0.981723  0.130105 -0.142  0.88725    
## site102             0.171078  1.186583  0.173472  0.986  0.32403    
## site103             0.545463  1.725407  0.123730  4.408 1.04e-05 ***
## site104             0.383722  1.467737  0.139757  2.746  0.00604 ** 
## site105             0.084237  1.087887  0.129388  0.651  0.51502    
## site106             0.095661  1.100386  0.121294  0.789  0.43030    
## site107             0.047112  1.048240  0.204823  0.230  0.81808    
## site108            -0.420855  0.656485  0.282620 -1.489  0.13646    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SS                    0.8550     1.1695    0.3164     2.311
## dx_yr                 2.2559     0.4433    2.1167     2.404
## age_dx                1.0121     0.9881    1.0032     1.021
## sexF                  0.9547     1.0475    0.8207     1.110
## dich_RaceNon-White    0.8695     1.1501    0.7005     1.079
## smokeHxFormer         1.0089     0.9912    0.8602     1.183
## smokeHxAlways         1.0548     0.9480    0.7578     1.468
## smokeHxUnknown        1.4939     0.6694    0.3540     6.304
## disadv                0.9817     1.0186    0.7608     1.267
## site102               1.1866     0.8428    0.8446     1.667
## site103               1.7254     0.5796    1.3539     2.199
## site104               1.4677     0.6813    1.1161     1.930
## site105               1.0879     0.9192    0.8442     1.402
## site106               1.1004     0.9088    0.8676     1.396
## site107               1.0482     0.9540    0.7016     1.566
## site108               0.6565     1.5233    0.3773     1.142
## 
## Concordance= 0.776  (se = 0.01 )
## Likelihood ratio test= 1063  on 16 df,   p=<2e-16
## Wald test            = 672.9  on 16 df,   p=<2e-16
## Score (logrank) test = 594.7  on 16 df,   p=<2e-16

91.7.4 IPF-Only SS - Combined Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ SS + dx_yr + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS + dx_yr + site, 
##     data = All_IPF, id = ID, cluster = cohort)
## 
##   n= 123367, number of events= 2749 
##    (723 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## SS       0.030974  1.031458  0.125277  0.162608   0.190 0.848933    
## dx_yr    0.103946  1.109540  0.006944  0.076371   1.361 0.173492    
## site02R  0.630156  1.877903  0.368953  0.125183   5.034 4.81e-07 ***
## site03R -0.269560  0.763715  0.347486  0.045283  -5.953 2.64e-09 ***
## site04R -0.223640  0.799603  0.449486  0.094828  -2.358 0.018355 *  
## site05R -0.323762  0.723422  0.368899  0.033701  -9.607  < 2e-16 ***
## site06R -0.040709  0.960109  0.325358  0.048845  -0.833 0.404604    
## site07R -0.321802  0.724842  0.304787  0.108441  -2.968 0.003002 ** 
## site09R -0.038104  0.962613  0.339958  0.015775  -2.415 0.015717 *  
## site1    0.247471  1.280782  0.281355  0.091382   2.708 0.006767 ** 
## site101 -0.107242  0.898308  0.293348  0.150615  -0.712 0.476447    
## site102 -0.230252  0.794333  0.299009  0.146218  -1.575 0.115323    
## site103  0.095934  1.100687  0.286336  0.081748   1.174 0.240579    
## site104 -0.188759  0.827986  0.294693  0.089934  -2.099 0.035829 *  
## site105  0.086678  1.090545  0.289693  0.158618   0.546 0.584752    
## site106 -0.207888  0.812298  0.286322  0.102916  -2.020 0.043385 *  
## site107  0.216668  1.241931  0.326836  0.168600   1.285 0.198759    
## site108  0.156595  1.169521  0.370614  0.245100   0.639 0.522888    
## site10R -0.175954  0.838657  0.398344  0.062455  -2.817 0.004843 ** 
## site11R -0.059040  0.942669  0.299373  0.064684  -0.913 0.361381    
## site12R -0.096060  0.908410  0.334443  0.013505  -7.113 1.14e-12 ***
## site13R -0.296467  0.743440  0.301832  0.105809  -2.802 0.005080 ** 
## site14R -0.367835  0.692232  0.640763  0.125490  -2.931 0.003377 ** 
## site15R  0.030171  1.030631  0.331102  0.015587   1.936 0.052906 .  
## site16R  0.136186  1.145895  0.344908  0.054401   2.503 0.012301 *  
## site17R  0.046209  1.047293  0.360297  0.028774   1.606 0.108294    
## site18R -0.266810  0.765819  0.322633  0.126094  -2.116 0.034348 *  
## site19R -0.029385  0.971043  0.350014  0.021198  -1.386 0.165677    
## site20R -0.268295  0.764682  0.347589  0.035996  -7.454 9.09e-14 ***
## site21R -0.309821  0.733578  0.319424  0.057242  -5.412 6.22e-08 ***
## site22R -0.072981  0.929618  0.305360  0.014113  -5.171 2.33e-07 ***
## site23R -0.164354  0.848441  0.322446  0.017296  -9.503  < 2e-16 ***
## site24R -0.161317  0.851023  0.331053  0.024477  -6.591 4.38e-11 ***
## site25R -0.056907  0.944682  0.354383  0.137682  -0.413 0.679373    
## site26R -0.431052  0.649825  0.351739  0.024409 -17.660  < 2e-16 ***
## site27R -0.013442  0.986648  0.759717  0.124104  -0.108 0.913746    
## site28R -0.137851  0.871229  0.433885  0.024705  -5.580 2.41e-08 ***
## site29R -0.316482  0.728708  0.435915  0.028476 -11.114  < 2e-16 ***
## site30R -0.211492  0.809376  0.330551  0.103844  -2.037 0.041687 *  
## site31R -0.331768  0.717654  0.449720  0.091177  -3.639 0.000274 ***
## site32R -0.352554  0.702890  0.368485  0.038328  -9.198  < 2e-16 ***
## site33R -0.360796  0.697121  0.334123  0.044809  -8.052 8.15e-16 ***
## site34R -0.158377  0.853528  0.313520  0.054213  -2.921 0.003485 ** 
## site35R -0.117677  0.888983  0.333941  0.028404  -4.143 3.43e-05 ***
## site36R -0.217377  0.804627  0.337792  0.027174  -7.999 1.25e-15 ***
## site37R -0.182436  0.833238  0.333886  0.054554  -3.344 0.000825 ***
## site38R -0.296089  0.743721  0.328953  0.038789  -7.633 2.29e-14 ***
## site39R -0.214578  0.806882  0.360560  0.041030  -5.230 1.70e-07 ***
## site40R  0.159445  1.172860  0.344707  0.055151   2.891 0.003839 ** 
## site41R -0.217005  0.804926  0.368974  0.020661 -10.503  < 2e-16 ***
## site42R  0.082940  1.086477  0.359965  0.024748   3.351 0.000804 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## SS         1.0315     0.9695    0.7500    1.4186
## dx_yr      1.1095     0.9013    0.9553    1.2887
## site02R    1.8779     0.5325    1.4693    2.4001
## site03R    0.7637     1.3094    0.6989    0.8346
## site04R    0.7996     1.2506    0.6640    0.9629
## site05R    0.7234     1.3823    0.6772    0.7728
## site06R    0.9601     1.0415    0.8725    1.0566
## site07R    0.7248     1.3796    0.5861    0.8965
## site09R    0.9626     1.0388    0.9333    0.9928
## site1      1.2808     0.7808    1.0708    1.5320
## site101    0.8983     1.1132    0.6687    1.2068
## site102    0.7943     1.2589    0.5964    1.0579
## site103    1.1007     0.9085    0.9377    1.2920
## site104    0.8280     1.2077    0.6942    0.9876
## site105    1.0905     0.9170    0.7991    1.4882
## site106    0.8123     1.2311    0.6639    0.9938
## site107    1.2419     0.8052    0.8925    1.7283
## site108    1.1695     0.8551    0.7234    1.8908
## site10R    0.8387     1.1924    0.7420    0.9479
## site11R    0.9427     1.0608    0.8304    1.0701
## site12R    0.9084     1.1008    0.8847    0.9328
## site13R    0.7434     1.3451    0.6042    0.9148
## site14R    0.6922     1.4446    0.5413    0.8853
## site15R    1.0306     0.9703    0.9996    1.0626
## site16R    1.1459     0.8727    1.0300    1.2748
## site17R    1.0473     0.9548    0.9899    1.1081
## site18R    0.7658     1.3058    0.5981    0.9805
## site19R    0.9710     1.0298    0.9315    1.0122
## site20R    0.7647     1.3077    0.7126    0.8206
## site21R    0.7336     1.3632    0.6557    0.8207
## site22R    0.9296     1.0757    0.9043    0.9557
## site23R    0.8484     1.1786    0.8202    0.8777
## site24R    0.8510     1.1751    0.8112    0.8928
## site25R    0.9447     1.0586    0.7213    1.2373
## site26R    0.6498     1.5389    0.6195    0.6817
## site27R    0.9866     1.0135    0.7736    1.2583
## site28R    0.8712     1.1478    0.8300    0.9145
## site29R    0.7287     1.3723    0.6892    0.7705
## site30R    0.8094     1.2355    0.6603    0.9921
## site31R    0.7177     1.3934    0.6002    0.8581
## site32R    0.7029     1.4227    0.6520    0.7577
## site33R    0.6971     1.4345    0.6385    0.7611
## site34R    0.8535     1.1716    0.7675    0.9492
## site35R    0.8890     1.1249    0.8408    0.9399
## site36R    0.8046     1.2428    0.7629    0.8486
## site37R    0.8332     1.2001    0.7487    0.9273
## site38R    0.7437     1.3446    0.6893    0.8025
## site39R    0.8069     1.2393    0.7445    0.8745
## site40R    1.1729     0.8526    1.0527    1.3067
## site41R    0.8049     1.2424    0.7730    0.8382
## site42R    1.0865     0.9204    1.0350    1.1405
## 
## Concordance= 0.596  (se = 0.044 )
## Likelihood ratio test= 347.8  on 51 df,   p=<2e-16
## Wald test            = 242.9  on 51 df,   p=<2e-16
## Score (logrank) test = 337.7  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ SS + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ SS + dx_yr + age_dx + 
##     sex + dich_Race + smokeHx + disadv + site, data = All_IPF, 
##     id = ID, cluster = cohort)
## 
##   n= 121241, number of events= 2709 
##    (2849 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## SS                  0.021623  1.021859  0.128121  0.160282   0.135 0.892684    
## dx_yr               0.100773  1.106026  0.007308  0.081613   1.235 0.216917    
## age_dx              0.004225  1.004234  0.002446  0.002173   1.944 0.051860 .  
## sexF               -0.158137  0.853733  0.044860  0.084978  -1.861 0.062755 .  
## dich_RaceNon-White  0.016703  1.016844  0.068774  0.056977   0.293 0.769401    
## smokeHxFormer       0.098784  1.103828  0.059392  0.036799   2.684 0.007265 ** 
## smokeHxAlways      -0.031082  0.969396  0.138366  0.116463  -0.267 0.789561    
## smokeHxUnknown      0.317872  1.374200  0.153713  0.231101   1.375 0.168986    
## smokeHxEver         0.036053  1.036711  0.064255  0.008216   4.388 1.14e-05 ***
## disadv              0.043051  1.043991  0.070452  0.130195   0.331 0.740898    
## site02R             0.629160  1.876034  0.380493  0.107660   5.844 5.10e-09 ***
## site03R            -0.249026  0.779560  0.357495  0.056157  -4.434 9.23e-06 ***
## site04R            -0.217899  0.804206  0.457644  0.115842  -1.881 0.059972 .  
## site05R            -0.320631  0.725691  0.379140  0.032248  -9.943  < 2e-16 ***
## site06R            -0.054131  0.947308  0.337840  0.075797  -0.714 0.475132    
## site07R            -0.329551  0.719246  0.316328  0.107299  -3.071 0.002131 ** 
## site09R            -0.084435  0.919031  0.352301  0.003247 -26.004  < 2e-16 ***
## site1               0.179764  1.196935  0.297461  0.129803   1.385 0.166083    
## site101            -0.158936  0.853051  0.308266  0.137811  -1.153 0.248791    
## site102            -0.301570  0.739656  0.314611  0.143872  -2.096 0.036073 *  
## site103             0.020367  1.020575  0.302002  0.047936   0.425 0.670930    
## site104            -0.254017  0.775679  0.310454  0.061098  -4.158 3.22e-05 ***
## site105             0.027700  1.028087  0.306629  0.143630   0.193 0.847073    
## site106            -0.286677  0.750754  0.302123  0.084328  -3.400 0.000675 ***
## site107             0.122928  1.130803  0.342294  0.144849   0.849 0.396068    
## site108             0.080164  1.083465  0.383060  0.225771   0.355 0.722538    
## site10R            -0.233663  0.791629  0.410039  0.084048  -2.780 0.005434 ** 
## site11R            -0.092688  0.911478  0.313633  0.055257  -1.677 0.093462 .  
## site12R            -0.117967  0.888726  0.345460  0.007065 -16.696  < 2e-16 ***
## site13R            -0.334295  0.715843  0.313297  0.107842  -3.100 0.001936 ** 
## site14R            -0.373236  0.688503  0.646458  0.095411  -3.912 9.16e-05 ***
## site15R            -0.008695  0.991343  0.343390  0.036035  -0.241 0.809327    
## site16R             0.141014  1.151441  0.355026  0.059606   2.366 0.017992 *  
## site17R             0.031445  1.031945  0.370900  0.032133   0.979 0.327786    
## site18R            -0.295970  0.743810  0.334789  0.101018  -2.930 0.003391 ** 
## site19R            -0.072056  0.930479  0.360650  0.016226  -4.441 8.97e-06 ***
## site20R            -0.266286  0.766220  0.358673  0.039675  -6.712 1.92e-11 ***
## site21R            -0.348036  0.706073  0.330695  0.046107  -7.548 4.41e-14 ***
## site22R            -0.082694  0.920633  0.317269  0.007929 -10.429  < 2e-16 ***
## site23R            -0.221899  0.800996  0.336102  0.012504 -17.746  < 2e-16 ***
## site24R            -0.174455  0.839915  0.342537  0.023248  -7.504 6.18e-14 ***
## site25R            -0.061254  0.940584  0.364842  0.105983  -0.578 0.563290    
## site26R            -0.465184  0.628020  0.363114  0.040173 -11.580  < 2e-16 ***
## site27R             0.002709  1.002712  0.766945  0.141578   0.019 0.984737    
## site28R            -0.138738  0.870456  0.442462  0.012563 -11.044  < 2e-16 ***
## site29R            -0.317423  0.728023  0.445291  0.026340 -12.051  < 2e-16 ***
## site30R            -0.274692  0.759806  0.342643  0.097338  -2.822 0.004772 ** 
## site31R            -0.368158  0.692008  0.457807  0.083588  -4.404 1.06e-05 ***
## site32R            -0.394230  0.674199  0.388704  0.030653 -12.861  < 2e-16 ***
## site33R            -0.398340  0.671434  0.347480  0.021494 -18.532  < 2e-16 ***
## site34R            -0.173392  0.840808  0.326115  0.054937  -3.156 0.001598 ** 
## site35R            -0.167913  0.845427  0.347040  0.016211 -10.358  < 2e-16 ***
## site36R            -0.223628  0.799613  0.348817  0.017914 -12.483  < 2e-16 ***
## site37R            -0.198711  0.819787  0.344636  0.061539  -3.229 0.001242 ** 
## site38R            -0.302839  0.738718  0.340147  0.019941 -15.186  < 2e-16 ***
## site39R            -0.257711  0.772819  0.370886  0.029439  -8.754  < 2e-16 ***
## site40R             0.119826  1.127300  0.354894  0.030382   3.944 8.02e-05 ***
## site41R            -0.254612  0.775217  0.378313  0.013613 -18.704  < 2e-16 ***
## site42R             0.058738  1.060497  0.370001  0.034405   1.707 0.087774 .  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## SS                    1.0219     0.9786    0.7464    1.3990
## dx_yr                 1.1060     0.9041    0.9425    1.2979
## age_dx                1.0042     0.9958    1.0000    1.0085
## sexF                  0.8537     1.1713    0.7228    1.0085
## dich_RaceNon-White    1.0168     0.9834    0.9094    1.1370
## smokeHxFormer         1.1038     0.9059    1.0270    1.1864
## smokeHxAlways         0.9694     1.0316    0.7716    1.2180
## smokeHxUnknown        1.3742     0.7277    0.8736    2.1615
## smokeHxEver           1.0367     0.9646    1.0202    1.0535
## disadv                1.0440     0.9579    0.8089    1.3475
## site02R               1.8760     0.5330    1.5191    2.3168
## site03R               0.7796     1.2828    0.6983    0.8703
## site04R               0.8042     1.2435    0.6409    1.0092
## site05R               0.7257     1.3780    0.6812    0.7730
## site06R               0.9473     1.0556    0.8165    1.0990
## site07R               0.7192     1.3903    0.5828    0.8876
## site09R               0.9190     1.0881    0.9132    0.9249
## site1                 1.1969     0.8355    0.9281    1.5437
## site101               0.8531     1.1723    0.6511    1.1176
## site102               0.7397     1.3520    0.5579    0.9806
## site103               1.0206     0.9798    0.9291    1.1211
## site104               0.7757     1.2892    0.6881    0.8744
## site105               1.0281     0.9727    0.7758    1.3623
## site106               0.7508     1.3320    0.6364    0.8857
## site107               1.1308     0.8843    0.8513    1.5020
## site108               1.0835     0.9230    0.6960    1.6865
## site10R               0.7916     1.2632    0.6714    0.9334
## site11R               0.9115     1.0971    0.8179    1.0157
## site12R               0.8887     1.1252    0.8765    0.9011
## site13R               0.7158     1.3970    0.5795    0.8843
## site14R               0.6885     1.4524    0.5711    0.8301
## site15R               0.9913     1.0087    0.9237    1.0639
## site16R               1.1514     0.8685    1.0245    1.2941
## site17R               1.0319     0.9690    0.9690    1.0990
## site18R               0.7438     1.3444    0.6102    0.9067
## site19R               0.9305     1.0747    0.9014    0.9605
## site20R               0.7662     1.3051    0.7089    0.8282
## site21R               0.7061     1.4163    0.6451    0.7729
## site22R               0.9206     1.0862    0.9064    0.9351
## site23R               0.8010     1.2484    0.7816    0.8209
## site24R               0.8399     1.1906    0.8025    0.8791
## site25R               0.9406     1.0632    0.7642    1.1577
## site26R               0.6280     1.5923    0.5805    0.6795
## site27R               1.0027     0.9973    0.7597    1.3234
## site28R               0.8705     1.1488    0.8493    0.8922
## site29R               0.7280     1.3736    0.6914    0.7666
## site30R               0.7598     1.3161    0.6278    0.9195
## site31R               0.6920     1.4451    0.5874    0.8152
## site32R               0.6742     1.4832    0.6349    0.7159
## site33R               0.6714     1.4894    0.6437    0.7003
## site34R               0.8408     1.1893    0.7550    0.9364
## site35R               0.8454     1.1828    0.8190    0.8727
## site36R               0.7996     1.2506    0.7720    0.8282
## site37R               0.8198     1.2198    0.7266    0.9249
## site38R               0.7387     1.3537    0.7104    0.7682
## site39R               0.7728     1.2940    0.7295    0.8187
## site40R               1.1273     0.8871    1.0621    1.1965
## site41R               0.7752     1.2900    0.7548    0.7962
## site42R               1.0605     0.9430    0.9913    1.1345
## 
## Concordance= 0.598  (se = 0.039 )
## Likelihood ratio test= 379.9  on 59 df,   p=<2e-16
## Wald test            = 880  on 59 df,   p=<2e-16
## Score (logrank) test = 374.6  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

91.8 IPF-Only Soil

91.8.1 IPF-Only Soil - Simmons Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ Soil + dx_yr, data=Simm_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil + dx_yr, 
##     data = Simm_IPF, id = ID)
## 
##   n= 31982, number of events= 695 
##    (446 observations deleted due to missingness)
## 
##            coef exp(coef)  se(coef)      z Pr(>|z|)   
## Soil  -0.063871  0.938126  0.280443 -0.228  0.81984   
## dx_yr  0.023282  1.023555  0.008582  2.713  0.00667 **
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##       exp(coef) exp(-coef) lower .95 upper .95
## Soil     0.9381      1.066    0.5414     1.625
## dx_yr    1.0236      0.977    1.0065     1.041
## 
## Concordance= 0.521  (se = 0.012 )
## Likelihood ratio test= 7.64  on 2 df,   p=0.02
## Wald test            = 7.62  on 2 df,   p=0.02
## Score (logrank) test = 7.64  on 2 df,   p=0.02

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ Soil + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv, data=Simm_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv, data = Simm_IPF, 
##     id = ID)
## 
##   n= 30488, number of events= 670 
##    (1940 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## Soil               -0.171177  0.842672  0.285003 -0.601  0.54810    
## dx_yr               0.010725  1.010783  0.009161  1.171  0.24169    
## age_dx              0.002084  1.002086  0.004566  0.456  0.64817    
## sexF               -0.385625  0.680026  0.088824 -4.341 1.42e-05 ***
## dich_RaceNon-White  0.157913  1.171065  0.128320  1.231  0.21846    
## smokeHxFormer       0.090807  1.095058  0.091966  0.987  0.32345    
## smokeHxAlways      -0.423249  0.654916  0.266943 -1.586  0.11284    
## smokeHxUnknown      0.514343  1.672540  0.170205  3.022  0.00251 ** 
## disadv              0.375604  1.455871  0.133916  2.805  0.00504 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## Soil                  0.8427     1.1867    0.4820    1.4732
## dx_yr                 1.0108     0.9893    0.9928    1.0291
## age_dx                1.0021     0.9979    0.9932    1.0111
## sexF                  0.6800     1.4705    0.5714    0.8093
## dich_RaceNon-White    1.1711     0.8539    0.9107    1.5059
## smokeHxFormer         1.0951     0.9132    0.9144    1.3113
## smokeHxAlways         0.6549     1.5269    0.3881    1.1051
## smokeHxUnknown        1.6725     0.5979    1.1981    2.3348
## disadv                1.4559     0.6869    1.1198    1.8928
## 
## Concordance= 0.591  (se = 0.012 )
## Likelihood ratio test= 51.98  on 9 df,   p=5e-08
## Wald test            = 52.49  on 9 df,   p=4e-08
## Score (logrank) test = 53.21  on 9 df,   p=3e-08

91.8.2 IPF-Only Soil - PFF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ Soil + dx_yr + site, data=PFF_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil + dx_yr + 
##     site, data = PFF_IPF, id = ID)
## 
##   n= 54499, number of events= 1146 
##    (217 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef)      z Pr(>|z|)    
## Soil    -0.13733   0.87168  0.19775 -0.694   0.4874    
## dx_yr    0.10437   1.11001  0.01355  7.700 1.37e-14 ***
## site02R  0.68139   1.97661  0.33727  2.020   0.0434 *  
## site03R -0.15200   0.85899  0.37177 -0.409   0.6826    
## site04R -0.17379   0.84048  0.45210 -0.384   0.7007    
## site05R -0.26063   0.77057  0.37271 -0.699   0.4844    
## site06R  0.01516   1.01528  0.32954  0.046   0.9633    
## site07R -0.26505   0.76717  0.30966 -0.856   0.3920    
## site08R       NA        NA  0.00000     NA       NA    
## site09R -0.01294   0.98714  0.33974 -0.038   0.9696    
## site10R -0.04774   0.95339  0.41012 -0.116   0.9073    
## site11R  0.01097   1.01103  0.31503  0.035   0.9722    
## site12R -0.05919   0.94253  0.33532 -0.177   0.8599    
## site13R -0.27381   0.76048  0.30251 -0.905   0.3654    
## site14R -0.34498   0.70823  0.64120 -0.538   0.5906    
## site15R  0.07651   1.07951  0.33385  0.229   0.8187    
## site16R  0.15493   1.16758  0.34484  0.449   0.6532    
## site17R  0.09730   1.10219  0.36834  0.264   0.7917    
## site18R -0.18592   0.83034  0.34179 -0.544   0.5865    
## site19R  0.15499   1.16765  0.42179  0.367   0.7133    
## site20R -0.18301   0.83276  0.35775 -0.512   0.6090    
## site21R -0.22577   0.79791  0.32849 -0.687   0.4919    
## site22R -0.04354   0.95739  0.30546 -0.143   0.8866    
## site23R -0.03103   0.96944  0.36591 -0.085   0.9324    
## site24R -0.11852   0.88823  0.33176 -0.357   0.7209    
## site25R  0.01875   1.01892  0.32670  0.057   0.9542    
## site26R -0.26606   0.76640  0.39041 -0.681   0.4956    
## site27R  0.02740   1.02778  0.75985  0.036   0.9712    
## site28R -0.08049   0.92266  0.43583 -0.185   0.8535    
## site29R -0.27336   0.76082  0.43374 -0.630   0.5285    
## site30R -0.15047   0.86030  0.33782 -0.445   0.6560    
## site31R -0.30214   0.73924  0.44963 -0.672   0.5016    
## site32R -0.10390   0.90132  0.50199 -0.207   0.8360    
## site33R -0.19395   0.82370  0.38479 -0.504   0.6142    
## site34R -0.13289   0.87556  0.31355 -0.424   0.6717    
## site35R -0.11603   0.89045  0.33429 -0.347   0.7285    
## site36R -0.15534   0.85612  0.34128 -0.455   0.6490    
## site37R -0.14982   0.86086  0.33414 -0.448   0.6539    
## site38R -0.26203   0.76949  0.32968 -0.795   0.4267    
## site39R -0.12656   0.88112  0.37466 -0.338   0.7355    
## site40R  0.27915   1.32201  0.37468  0.745   0.4562    
## site41R -0.17072   0.84305  0.37054 -0.461   0.6450    
## site42R  0.09619   1.10097  0.36030  0.267   0.7895    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## Soil       0.8717     1.1472    0.5916     1.284
## dx_yr      1.1100     0.9009    1.0809     1.140
## site02R    1.9766     0.5059    1.0206     3.828
## site03R    0.8590     1.1642    0.4145     1.780
## site04R    0.8405     1.1898    0.3465     2.039
## site05R    0.7706     1.2977    0.3712     1.600
## site06R    1.0153     0.9850    0.5322     1.937
## site07R    0.7672     1.3035    0.4181     1.408
## site08R        NA         NA        NA        NA
## site09R    0.9871     1.0130    0.5072     1.921
## site10R    0.9534     1.0489    0.4268     2.130
## site11R    1.0110     0.9891    0.5453     1.875
## site12R    0.9425     1.0610    0.4885     1.819
## site13R    0.7605     1.3150    0.4203     1.376
## site14R    0.7082     1.4120    0.2015     2.489
## site15R    1.0795     0.9263    0.5611     2.077
## site16R    1.1676     0.8565    0.5940     2.295
## site17R    1.1022     0.9073    0.5355     2.269
## site18R    0.8303     1.2043    0.4249     1.623
## site19R    1.1676     0.8564    0.5108     2.669
## site20R    0.8328     1.2008    0.4130     1.679
## site21R    0.7979     1.2533    0.4191     1.519
## site22R    0.9574     1.0445    0.5261     1.742
## site23R    0.9694     1.0315    0.4732     1.986
## site24R    0.8882     1.1258    0.4636     1.702
## site25R    1.0189     0.9814    0.5371     1.933
## site26R    0.7664     1.3048    0.3566     1.647
## site27R    1.0278     0.9730    0.2318     4.557
## site28R    0.9227     1.0838    0.3927     2.168
## site29R    0.7608     1.3144    0.3251     1.780
## site30R    0.8603     1.1624    0.4437     1.668
## site31R    0.7392     1.3527    0.3062     1.784
## site32R    0.9013     1.1095    0.3370     2.411
## site33R    0.8237     1.2140    0.3875     1.751
## site34R    0.8756     1.1421    0.4736     1.619
## site35R    0.8904     1.1230    0.4624     1.715
## site36R    0.8561     1.1681    0.4386     1.671
## site37R    0.8609     1.1616    0.4472     1.657
## site38R    0.7695     1.2996    0.4033     1.468
## site39R    0.8811     1.1349    0.4228     1.836
## site40R    1.3220     0.7564    0.6343     2.755
## site41R    0.8431     1.1862    0.4078     1.743
## site42R    1.1010     0.9083    0.5434     2.231
## 
## Concordance= 0.59  (se = 0.009 )
## Likelihood ratio test= 111  on 42 df,   p=4e-08
## Wald test            = 109.2  on 42 df,   p=7e-08
## Score (logrank) test = 113.5  on 42 df,   p=2e-08

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ Soil + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=PFF_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = PFF_IPF, 
##     id = ID)
## 
##   n= 53867, number of events= 1131 
##    (849 observations deleted due to missingness)
## 
##                          coef  exp(coef)   se(coef)      z Pr(>|z|)    
## Soil               -0.0934647  0.9107701  0.2028037 -0.461   0.6449    
## dx_yr               0.1029613  1.1084485  0.0139904  7.359 1.85e-13 ***
## age_dx             -0.0003036  0.9996964  0.0039352 -0.077   0.9385    
## sexM                0.0616450  1.0635847  0.0713035  0.865   0.3873    
## dich_RaceNon-White -0.0553269  0.9461758  0.1260294 -0.439   0.6607    
## smokeHxEver         0.0432725  1.0442224  0.0646213  0.670   0.5031    
## disadv             -0.0468679  0.9542134  0.1128079 -0.415   0.6778    
## site02R             0.6313496  1.8801463  0.3495795  1.806   0.0709 .  
## site03R            -0.2051793  0.8145012  0.3855647 -0.532   0.5946    
## site04R            -0.2309114  0.7938098  0.4626797 -0.499   0.6177    
## site05R            -0.3102742  0.7332459  0.3850124 -0.806   0.4203    
## site06R            -0.0277814  0.9726009  0.3448652 -0.081   0.9358    
## site07R            -0.3132185  0.7310901  0.3238172 -0.967   0.3334    
## site08R                    NA         NA  0.0000000     NA       NA    
## site09R            -0.0859959  0.9175980  0.3538756 -0.243   0.8080    
## site10R            -0.1303326  0.8778035  0.4247409 -0.307   0.7590    
## site11R            -0.0612369  0.9406004  0.3330850 -0.184   0.8541    
## site12R            -0.1040231  0.9012045  0.3491777 -0.298   0.7658    
## site13R            -0.3273849  0.7208062  0.3155908 -1.037   0.2996    
## site14R            -0.3983486  0.6714279  0.6482392 -0.615   0.5389    
## site15R             0.0234922  1.0237703  0.3492368  0.067   0.9464    
## site16R             0.1127367  1.1193372  0.3560721  0.317   0.7515    
## site17R             0.0297750  1.0302227  0.3819637  0.078   0.9379    
## site18R            -0.2395566  0.7869767  0.3572108 -0.671   0.5025    
## site19R             0.0616245  1.0635629  0.4376932  0.141   0.8880    
## site20R            -0.2468888  0.7812276  0.3722964 -0.663   0.5072    
## site21R            -0.2793362  0.7562856  0.3423697 -0.816   0.4146    
## site22R            -0.0925218  0.9116293  0.3193501 -0.290   0.7720    
## site23R            -0.1267154  0.8809844  0.3827611 -0.331   0.7406    
## site24R            -0.1727503  0.8413477  0.3454764 -0.500   0.6171    
## site25R            -0.0139625  0.9861346  0.3399094 -0.041   0.9672    
## site26R            -0.3352965  0.7151260  0.4042842 -0.829   0.4069    
## site27R            -0.0712275  0.9312500  0.7705994 -0.092   0.9264    
## site28R            -0.1142890  0.8920001  0.4470131 -0.256   0.7982    
## site29R            -0.3194236  0.7265677  0.4448285 -0.718   0.4727    
## site30R            -0.2134605  0.8077841  0.3512371 -0.608   0.5434    
## site31R            -0.3485804  0.7056892  0.4585690 -0.760   0.4472    
## site32R            -0.2275125  0.7965125  0.5331045 -0.427   0.6695    
## site33R            -0.2733910  0.7607952  0.4005499 -0.683   0.4949    
## site34R            -0.1706071  0.8431528  0.3285899 -0.519   0.6036    
## site35R            -0.1655837  0.8473989  0.3481936 -0.476   0.6344    
## site36R            -0.1918885  0.8253989  0.3554542 -0.540   0.5893    
## site37R            -0.1890942  0.8277085  0.3465908 -0.546   0.5854    
## site38R            -0.2985583  0.7418870  0.3429988 -0.870   0.3841    
## site39R            -0.1944922  0.8232526  0.3881310 -0.501   0.6163    
## site40R             0.2195829  1.2455571  0.3863876  0.568   0.5698    
## site41R            -0.2117426  0.8091729  0.3808945 -0.556   0.5783    
## site42R             0.0349380  1.0355555  0.3721420  0.094   0.9252    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## Soil                  0.9108     1.0980    0.6120     1.355
## dx_yr                 1.1084     0.9022    1.0785     1.139
## age_dx                0.9997     1.0003    0.9920     1.007
## sexM                  1.0636     0.9402    0.9249     1.223
## dich_RaceNon-White    0.9462     1.0569    0.7391     1.211
## smokeHxEver           1.0442     0.9577    0.9200     1.185
## disadv                0.9542     1.0480    0.7649     1.190
## site02R               1.8801     0.5319    0.9476     3.730
## site03R               0.8145     1.2277    0.3826     1.734
## site04R               0.7938     1.2597    0.3205     1.966
## site05R               0.7332     1.3638    0.3448     1.559
## site06R               0.9726     1.0282    0.4947     1.912
## site07R               0.7311     1.3678    0.3876     1.379
## site08R                   NA         NA        NA        NA
## site09R               0.9176     1.0898    0.4586     1.836
## site10R               0.8778     1.1392    0.3818     2.018
## site11R               0.9406     1.0632    0.4896     1.807
## site12R               0.9012     1.1096    0.4546     1.787
## site13R               0.7208     1.3873    0.3883     1.338
## site14R               0.6714     1.4894    0.1885     2.392
## site15R               1.0238     0.9768    0.5163     2.030
## site16R               1.1193     0.8934    0.5570     2.249
## site17R               1.0302     0.9707    0.4873     2.178
## site18R               0.7870     1.2707    0.3908     1.585
## site19R               1.0636     0.9402    0.4510     2.508
## site20R               0.7812     1.2800    0.3766     1.621
## site21R               0.7563     1.3223    0.3866     1.479
## site22R               0.9116     1.0969    0.4875     1.705
## site23R               0.8810     1.1351    0.4161     1.865
## site24R               0.8413     1.1886    0.4275     1.656
## site25R               0.9861     1.0141    0.5065     1.920
## site26R               0.7151     1.3984    0.3238     1.579
## site27R               0.9312     1.0738    0.2057     4.217
## site28R               0.8920     1.1211    0.3714     2.142
## site29R               0.7266     1.3763    0.3038     1.737
## site30R               0.8078     1.2380    0.4058     1.608
## site31R               0.7057     1.4171    0.2873     1.734
## site32R               0.7965     1.2555    0.2802     2.264
## site33R               0.7608     1.3144    0.3470     1.668
## site34R               0.8432     1.1860    0.4428     1.605
## site35R               0.8474     1.1801    0.4283     1.677
## site36R               0.8254     1.2115    0.4112     1.657
## site37R               0.8277     1.2082    0.4196     1.633
## site38R               0.7419     1.3479    0.3788     1.453
## site39R               0.8233     1.2147    0.3847     1.762
## site40R               1.2456     0.8029    0.5841     2.656
## site41R               0.8092     1.2358    0.3836     1.707
## site42R               1.0356     0.9657    0.4994     2.148
## 
## Concordance= 0.591  (se = 0.009 )
## Likelihood ratio test= 111.4  on 47 df,   p=4e-07
## Wald test            = 109.9  on 47 df,   p=6e-07
## Score (logrank) test = 114.3  on 47 df,   p=2e-07

91.8.3 IPF-Only Soil - CARE-PF Cohort

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ Soil + dx_yr + site, data=CARE_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil + dx_yr + 
##     site, data = CARE_IPF, id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##             coef exp(coef) se(coef)      z Pr(>|z|)    
## Soil    -0.41584   0.65979  0.24801 -1.677  0.09360 .  
## dx_yr    0.81730   2.26437  0.03233 25.282  < 2e-16 ***
## site102  0.17519   1.19148  0.14943  1.172  0.24102    
## site103  0.41989   1.52179  0.13316  3.153  0.00161 ** 
## site104  0.28207   1.32587  0.14910  1.892  0.05851 .  
## site105  0.01151   1.01157  0.12992  0.089  0.92943    
## site106  0.16798   1.18291  0.12222  1.374  0.16931    
## site107  0.05400   1.05549  0.19968  0.270  0.78681    
## site108 -0.42330   0.65488  0.26383 -1.604  0.10861    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## Soil       0.6598     1.5156    0.4058     1.073
## dx_yr      2.2644     0.4416    2.1254     2.412
## site102    1.1915     0.8393    0.8890     1.597
## site103    1.5218     0.6571    1.1722     1.976
## site104    1.3259     0.7542    0.9899     1.776
## site105    1.0116     0.9886    0.7842     1.305
## site106    1.1829     0.8454    0.9309     1.503
## site107    1.0555     0.9474    0.7136     1.561
## site108    0.6549     1.5270    0.3905     1.098
## 
## Concordance= 0.774  (se = 0.01 )
## Likelihood ratio test= 1055  on 9 df,   p=<2e-16
## Wald test            = 669.2  on 9 df,   p=<2e-16
## Score (logrank) test = 590.4  on 9 df,   p=<2e-16

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ Soil + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site, data=CARE_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = CARE_IPF, 
##     id = ID)
## 
##   n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef)      z Pr(>|z|)    
## Soil               -0.520070  0.594479  0.255310 -2.037  0.04165 *  
## dx_yr               0.817926  2.265796  0.032507 25.162  < 2e-16 ***
## age_dx              0.012881  1.012965  0.004519  2.850  0.00437 ** 
## sexF               -0.042562  0.958331  0.077062 -0.552  0.58073    
## dich_RaceNon-White -0.142358  0.867311  0.110439 -1.289  0.19739    
## smokeHxFormer       0.004960  1.004972  0.081392  0.061  0.95141    
## smokeHxAlways       0.064492  1.066617  0.169049  0.381  0.70283    
## smokeHxUnknown      0.438159  1.549851  0.734863  0.596  0.55101    
## disadv              0.025556  1.025885  0.131460  0.194  0.84586    
## site102             0.149268  1.160985  0.151745  0.984  0.32527    
## site103             0.408201  1.504109  0.137222  2.975  0.00293 ** 
## site104             0.265885  1.304585  0.150802  1.763  0.07788 .  
## site105             0.015099  1.015214  0.133367  0.113  0.90986    
## site106             0.148186  1.159728  0.123647  1.198  0.23074    
## site107            -0.028714  0.971694  0.207966 -0.138  0.89018    
## site108            -0.432675  0.648771  0.267380 -1.618  0.10562    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## Soil                  0.5945     1.6821    0.3604    0.9805
## dx_yr                 2.2658     0.4413    2.1259    2.4149
## age_dx                1.0130     0.9872    1.0040    1.0220
## sexF                  0.9583     1.0435    0.8240    1.1146
## dich_RaceNon-White    0.8673     1.1530    0.6985    1.0769
## smokeHxFormer         1.0050     0.9951    0.8568    1.1788
## smokeHxAlways         1.0666     0.9375    0.7658    1.4856
## smokeHxUnknown        1.5499     0.6452    0.3671    6.5434
## disadv                1.0259     0.9748    0.7929    1.3274
## site102               1.1610     0.8613    0.8623    1.5631
## site103               1.5041     0.6648    1.1494    1.9683
## site104               1.3046     0.7665    0.9708    1.7532
## site105               1.0152     0.9850    0.7817    1.3185
## site106               1.1597     0.8623    0.9101    1.4778
## site107               0.9717     1.0291    0.6464    1.4607
## site108               0.6488     1.5414    0.3841    1.0957
## 
## Concordance= 0.776  (se = 0.01 )
## Likelihood ratio test= 1068  on 16 df,   p=<2e-16
## Wald test            = 677.4  on 16 df,   p=<2e-16
## Score (logrank) test = 596.7  on 16 df,   p=<2e-16

91.8.4 IPF-Only Soil - Combined Cohorts

coxPH_model1 <- coxph(Surv(start, end, event==1) ~ Soil + dx_yr + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model1)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil + dx_yr + 
##     site, data = All_IPF, id = ID, cluster = cohort)
## 
##   n= 123367, number of events= 2749 
##    (723 observations deleted due to missingness)
## 
##              coef exp(coef)  se(coef) robust se      z Pr(>|z|)    
## Soil     0.096461  1.101266  0.133647  0.133849  0.721 0.471116    
## dx_yr    0.104064  1.109671  0.006936  0.075734  1.374 0.169420    
## site02R  0.652506  1.920348  0.335113  0.078593  8.302  < 2e-16 ***
## site03R -0.340474  0.711433  0.359238  0.113012 -3.013 0.002589 ** 
## site04R -0.248483  0.779983  0.450553  0.096029 -2.588 0.009666 ** 
## site05R -0.354093  0.701809  0.370356  0.055673 -6.360 2.01e-10 ***
## site06R -0.070708  0.931734  0.326867  0.085224 -0.830 0.406723    
## site07R -0.349283  0.705194  0.307018  0.116545 -2.997 0.002727 ** 
## site09R -0.034744  0.965853  0.339721  0.019175 -1.812 0.069995 .  
## site1    0.225422  1.252851  0.282632  0.109365  2.061 0.039285 *  
## site101 -0.125572  0.881992  0.294226  0.145474 -0.863 0.388031    
## site102 -0.243851  0.783604  0.298041  0.113797 -2.143 0.032125 *  
## site103  0.103426  1.108964  0.286428  0.086177  1.200 0.230078    
## site104 -0.184790  0.831279  0.294742  0.088643 -2.085 0.037100 *  
## site105  0.080071  1.083364  0.289690  0.149090  0.537 0.591224    
## site106 -0.234705  0.790804  0.288592  0.108516 -2.163 0.030552 *  
## site107  0.210863  1.234743  0.326842  0.160289  1.316 0.188338    
## site108  0.138246  1.148258  0.369865  0.208500  0.663 0.507297    
## site10R -0.219847  0.802642  0.396887  0.122056 -1.801 0.071673 .  
## site11R -0.106694  0.898801  0.306454  0.093755 -1.138 0.255118    
## site12R -0.115310  0.891090  0.334434  0.037397 -3.083 0.002047 ** 
## site13R -0.305440  0.736799  0.301995  0.101017 -3.024 0.002498 ** 
## site14R -0.368476  0.691788  0.640729  0.116835 -3.154 0.001612 ** 
## site15R  0.012552  1.012631  0.331923  0.039339  0.319 0.749672    
## site16R  0.132608  1.141802  0.344574  0.032946  4.025 5.70e-05 ***
## site17R  0.004196  1.004205  0.363788  0.054377  0.077 0.938486    
## site18R -0.315793  0.729210  0.330579  0.176975 -1.784 0.074359 .  
## site19R -0.147884  0.862531  0.385297  0.176120 -0.840 0.401090    
## site20R -0.314603  0.730079  0.351839  0.077011 -4.085 4.40e-05 ***
## site21R -0.348625  0.705657  0.323476  0.085881 -4.059 4.92e-05 ***
## site22R -0.076702  0.926165  0.305238  0.021875 -3.506 0.000454 ***
## site23R -0.248213  0.780193  0.342939  0.126912 -1.956 0.050489 .  
## site24R -0.176336  0.838336  0.330958  0.052781 -3.341 0.000835 ***
## site25R -0.038912  0.961835  0.325702  0.059452 -0.655 0.512779    
## site26R -0.502636  0.604934  0.365918  0.160685 -3.128 0.001760 ** 
## site27R -0.018661  0.981512  0.759684  0.113857 -0.164 0.869809    
## site28R -0.161772  0.850635  0.434634  0.048203 -3.356 0.000791 ***
## site29R -0.305843  0.736502  0.433702  0.047544 -6.433 1.25e-10 ***
## site30R -0.244573  0.783039  0.333738  0.119179 -2.052 0.040155 *  
## site31R -0.340651  0.711307  0.449483  0.071989 -4.732 2.22e-06 ***
## site32R -0.520670  0.594122  0.436103  0.244481 -2.130 0.033196 *  
## site33R -0.446444  0.639900  0.357086  0.171583 -2.602 0.009271 ** 
## site34R -0.162833  0.849733  0.313460  0.045958 -3.543 0.000395 ***
## site35R -0.106809  0.898697  0.334019  0.037740 -2.830 0.004653 ** 
## site36R -0.243487  0.783889  0.339251  0.052750 -4.616 3.91e-06 ***
## site37R -0.189396  0.827459  0.333946  0.051262 -3.695 0.000220 ***
## site38R -0.306910  0.735716  0.329243  0.042347 -7.248 4.24e-13 ***
## site39R -0.270550  0.762959  0.366888  0.085432 -3.167 0.001541 ** 
## site40R  0.095661  1.100386  0.357236  0.153517  0.623 0.533197    
## site41R -0.231058  0.793693  0.369392  0.055471 -4.165 3.11e-05 ***
## site42R  0.076538  1.079544  0.360078  0.027902  2.743 0.006086 ** 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##         exp(coef) exp(-coef) lower .95 upper .95
## Soil       1.1013     0.9080    0.8471    1.4316
## dx_yr      1.1097     0.9012    0.9566    1.2872
## site02R    1.9203     0.5207    1.6462    2.2402
## site03R    0.7114     1.4056    0.5701    0.8878
## site04R    0.7800     1.2821    0.6462    0.9415
## site05R    0.7018     1.4249    0.6293    0.7827
## site06R    0.9317     1.0733    0.7884    1.1011
## site07R    0.7052     1.4180    0.5612    0.8862
## site09R    0.9659     1.0354    0.9302    1.0028
## site1      1.2529     0.7982    1.0111    1.5524
## site101    0.8820     1.1338    0.6632    1.1730
## site102    0.7836     1.2762    0.6269    0.9794
## site103    1.1090     0.9017    0.9366    1.3130
## site104    0.8313     1.2030    0.6987    0.9890
## site105    1.0834     0.9231    0.8089    1.4510
## site106    0.7908     1.2645    0.6393    0.9782
## site107    1.2347     0.8099    0.9019    1.6905
## site108    1.1483     0.8709    0.7631    1.7279
## site10R    0.8026     1.2459    0.6319    1.0196
## site11R    0.8988     1.1126    0.7479    1.0801
## site12R    0.8911     1.1222    0.8281    0.9589
## site13R    0.7368     1.3572    0.6045    0.8981
## site14R    0.6918     1.4455    0.5502    0.8698
## site15R    1.0126     0.9875    0.9375    1.0938
## site16R    1.1418     0.8758    1.0704    1.2180
## site17R    1.0042     0.9958    0.9027    1.1171
## site18R    0.7292     1.3713    0.5155    1.0316
## site19R    0.8625     1.1594    0.6107    1.2181
## site20R    0.7301     1.3697    0.6278    0.8490
## site21R    0.7057     1.4171    0.5963    0.8350
## site22R    0.9262     1.0797    0.8873    0.9667
## site23R    0.7802     1.2817    0.6084    1.0005
## site24R    0.8383     1.1928    0.7559    0.9297
## site25R    0.9618     1.0397    0.8560    1.0807
## site26R    0.6049     1.6531    0.4415    0.8289
## site27R    0.9815     1.0188    0.7852    1.2269
## site28R    0.8506     1.1756    0.7740    0.9349
## site29R    0.7365     1.3578    0.6710    0.8084
## site30R    0.7830     1.2771    0.6199    0.9891
## site31R    0.7113     1.4059    0.6177    0.8191
## site32R    0.5941     1.6832    0.3679    0.9594
## site33R    0.6399     1.5627    0.4572    0.8957
## site34R    0.8497     1.1768    0.7765    0.9298
## site35R    0.8987     1.1127    0.8346    0.9677
## site36R    0.7839     1.2757    0.7069    0.8693
## site37R    0.8275     1.2085    0.7484    0.9149
## site38R    0.7357     1.3592    0.6771    0.7994
## site39R    0.7630     1.3107    0.6453    0.9020
## site40R    1.1004     0.9088    0.8145    1.4867
## site41R    0.7937     1.2599    0.7119    0.8848
## site42R    1.0795     0.9263    1.0221    1.1402
## 
## Concordance= 0.595  (se = 0.043 )
## Likelihood ratio test= 348.3  on 51 df,   p=<2e-16
## Wald test            = 2.32  on 51 df,   p=1
## Score (logrank) test = 337.7  on 51 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

coxPH_model2 <- coxph(Surv(start, end, event==1) ~ Soil + dx_yr + age_dx + sex + dich_Race + smokeHx + disadv + site + cluster(cohort), data=All_IPF, id=ID)
summary(coxPH_model2)

## Call:
## coxph(formula = Surv(start, end, event == 1) ~ Soil + dx_yr + 
##     age_dx + sex + dich_Race + smokeHx + disadv + site, data = All_IPF, 
##     id = ID, cluster = cohort)
## 
##   n= 121241, number of events= 2709 
##    (2849 observations deleted due to missingness)
## 
##                         coef exp(coef)  se(coef) robust se       z Pr(>|z|)    
## Soil                0.106677  1.112575  0.136467  0.117733   0.906 0.364884    
## dx_yr               0.100900  1.106166  0.007300  0.080994   1.246 0.212852    
## age_dx              0.004224  1.004233  0.002446  0.002114   1.998 0.045750 *  
## sexF               -0.157481  0.854293  0.044865  0.084406  -1.866 0.062076 .  
## dich_RaceNon-White  0.016495  1.016632  0.068702  0.060435   0.273 0.784901    
## smokeHxFormer       0.098347  1.103346  0.059368  0.038470   2.556 0.010574 *  
## smokeHxAlways      -0.029778  0.970661  0.138353  0.111049  -0.268 0.788584    
## smokeHxUnknown      0.316017  1.371653  0.153680  0.235887   1.340 0.180344    
## smokeHxEver         0.036472  1.037145  0.064269  0.008234   4.430 9.44e-06 ***
## disadv              0.036568  1.037245  0.070853  0.129685   0.282 0.777961    
## site02R             0.637105  1.890998  0.345770  0.101831   6.256 3.94e-10 ***
## site03R            -0.326363  0.721543  0.369836  0.097033  -3.363 0.000770 ***
## site04R            -0.245931  0.781976  0.458933  0.119661  -2.055 0.039857 *  
## site05R            -0.353007  0.702572  0.380812  0.038970  -9.059  < 2e-16 ***
## site06R            -0.084278  0.919176  0.339367  0.092366  -0.912 0.361540    
## site07R            -0.359488  0.698033  0.318634  0.121674  -2.955 0.003131 ** 
## site09R            -0.083154  0.920209  0.351989  0.015139  -5.493 3.95e-08 ***
## site1               0.156513  1.169426  0.298746  0.134484   1.164 0.244504    
## site101            -0.178533  0.836497  0.309186  0.138464  -1.289 0.197267    
## site102            -0.314564  0.730107  0.313739  0.113777  -2.765 0.005697 ** 
## site103             0.028942  1.029365  0.302123  0.050852   0.569 0.569266    
## site104            -0.248756  0.779770  0.310535  0.060744  -4.095 4.22e-05 ***
## site105             0.021678  1.021914  0.306668  0.137299   0.158 0.874546    
## site106            -0.316017  0.729047  0.304411  0.097004  -3.258 0.001123 ** 
## site107             0.117805  1.125025  0.342336  0.139086   0.847 0.396997    
## site108             0.063543  1.065606  0.382243  0.193529   0.328 0.742655    
## site10R            -0.291765  0.746944  0.407767  0.115284  -2.531 0.011379 *  
## site11R            -0.144870  0.865135  0.320638  0.088383  -1.639 0.101188    
## site12R            -0.137166  0.871825  0.345621  0.027984  -4.902 9.51e-07 ***
## site13R            -0.344672  0.708452  0.313547  0.108239  -3.184 0.001451 ** 
## site14R            -0.377135  0.685824  0.646436  0.081436  -4.631 3.64e-06 ***
## site15R            -0.029217  0.971206  0.343951  0.030594  -0.955 0.339584    
## site16R             0.134074  1.143477  0.354646  0.027724   4.836 1.33e-06 ***
## site17R            -0.012730  0.987350  0.374477  0.037667  -0.338 0.735384    
## site18R            -0.351510  0.703625  0.342338  0.168828  -2.082 0.037338 *  
## site19R            -0.202751  0.816481  0.397030  0.155643  -1.303 0.192687    
## site20R            -0.316339  0.728812  0.363331  0.057677  -5.485 4.14e-08 ***
## site21R            -0.390065  0.677013  0.334837  0.080963  -4.818 1.45e-06 ***
## site22R            -0.089115  0.914740  0.317119  0.026983  -3.303 0.000958 ***
## site23R            -0.314688  0.730017  0.356663  0.110430  -2.850 0.004377 ** 
## site24R            -0.189903  0.827039  0.342612  0.041268  -4.602 4.19e-06 ***
## site25R            -0.057732  0.943903  0.337658  0.079963  -0.722 0.470304    
## site26R            -0.550175  0.576849  0.376581  0.150283  -3.661 0.000251 ***
## site27R            -0.004913  0.995099  0.766972  0.135581  -0.036 0.971093    
## site28R            -0.163375  0.849272  0.443301  0.032192  -5.075 3.88e-07 ***
## site29R            -0.311727  0.732181  0.442859  0.084305  -3.698 0.000218 ***
## site30R            -0.312130  0.731886  0.345998  0.124065  -2.516 0.011874 *  
## site31R            -0.376698  0.686123  0.457649  0.068061  -5.535 3.12e-08 ***
## site32R            -0.584303  0.557494  0.459060  0.225207  -2.595 0.009472 ** 
## site33R            -0.494475  0.609891  0.369430  0.158082  -3.128 0.001760 ** 
## site34R            -0.177892  0.837033  0.326118  0.049431  -3.599 0.000320 ***
## site35R            -0.157731  0.854080  0.346996  0.025635  -6.153 7.60e-10 ***
## site36R            -0.250990  0.778030  0.350342  0.040173  -6.248 4.17e-10 ***
## site37R            -0.206351  0.813547  0.344747  0.060862  -3.390 0.000698 ***
## site38R            -0.314105  0.730443  0.340452  0.026105 -12.032  < 2e-16 ***
## site39R            -0.317998  0.727604  0.377636  0.075814  -4.194 2.74e-05 ***
## site40R             0.045278  1.046318  0.367447  0.147314   0.307 0.758573    
## site41R            -0.272098  0.761780  0.378653  0.054037  -5.035 4.77e-07 ***
## site42R             0.050405  1.051697  0.370161  0.038905   1.296 0.195121    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
##                    exp(coef) exp(-coef) lower .95 upper .95
## Soil                  1.1126     0.8988    0.8833    1.4013
## dx_yr                 1.1062     0.9040    0.9438    1.2965
## age_dx                1.0042     0.9958    1.0001    1.0084
## sexF                  0.8543     1.1706    0.7240    1.0080
## dich_RaceNon-White    1.0166     0.9836    0.9031    1.1445
## smokeHxFormer         1.1033     0.9063    1.0232    1.1898
## smokeHxAlways         0.9707     1.0302    0.7808    1.2067
## smokeHxUnknown        1.3717     0.7290    0.8639    2.1779
## smokeHxEver           1.0371     0.9642    1.0205    1.0540
## disadv                1.0372     0.9641    0.8044    1.3374
## site02R               1.8910     0.5288    1.5489    2.3087
## site03R               0.7215     1.3859    0.5966    0.8727
## site04R               0.7820     1.2788    0.6185    0.9887
## site05R               0.7026     1.4233    0.6509    0.7583
## site06R               0.9192     1.0879    0.7670    1.1016
## site07R               0.6980     1.4326    0.5499    0.8860
## site09R               0.9202     1.0867    0.8933    0.9479
## site1                 1.1694     0.8551    0.8985    1.5221
## site101               0.8365     1.1955    0.6377    1.0973
## site102               0.7301     1.3697    0.5842    0.9125
## site103               1.0294     0.9715    0.9317    1.1372
## site104               0.7798     1.2824    0.6922    0.8784
## site105               1.0219     0.9786    0.7808    1.3375
## site106               0.7290     1.3717    0.6028    0.8817
## site107               1.1250     0.8889    0.8566    1.4776
## site108               1.0656     0.9384    0.7292    1.5571
## site10R               0.7469     1.3388    0.5959    0.9363
## site11R               0.8651     1.1559    0.7275    1.0288
## site12R               0.8718     1.1470    0.8253    0.9210
## site13R               0.7085     1.4115    0.5730    0.8759
## site14R               0.6858     1.4581    0.5846    0.8045
## site15R               0.9712     1.0296    0.9147    1.0312
## site16R               1.1435     0.8745    1.0830    1.2073
## site17R               0.9874     1.0128    0.9171    1.0630
## site18R               0.7036     1.4212    0.5054    0.9796
## site19R               0.8165     1.2248    0.6018    1.1077
## site20R               0.7288     1.3721    0.6509    0.8160
## site21R               0.6770     1.4771    0.5777    0.7934
## site22R               0.9147     1.0932    0.8676    0.9644
## site23R               0.7300     1.3698    0.5879    0.9064
## site24R               0.8270     1.2091    0.7628    0.8967
## site25R               0.9439     1.0594    0.8070    1.1041
## site26R               0.5768     1.7336    0.4297    0.7744
## site27R               0.9951     1.0049    0.7629    1.2980
## site28R               0.8493     1.1775    0.7973    0.9046
## site29R               0.7322     1.3658    0.6207    0.8637
## site30R               0.7319     1.3663    0.5739    0.9334
## site31R               0.6861     1.4575    0.6004    0.7840
## site32R               0.5575     1.7937    0.3585    0.8668
## site33R               0.6099     1.6396    0.4474    0.8314
## site34R               0.8370     1.1947    0.7597    0.9222
## site35R               0.8541     1.1709    0.8122    0.8981
## site36R               0.7780     1.2853    0.7191    0.8418
## site37R               0.8135     1.2292    0.7221    0.9166
## site38R               0.7304     1.3690    0.6940    0.7688
## site39R               0.7276     1.3744    0.6271    0.8442
## site40R               1.0463     0.9557    0.7839    1.3966
## site41R               0.7618     1.3127    0.6852    0.8469
## site42R               1.0517     0.9508    0.9745    1.1350
## 
## Concordance= 0.598  (se = 0.039 )
## Likelihood ratio test= 380.5  on 59 df,   p=<2e-16
## Wald test            = 1.84  on 59 df,   p=1
## Score (logrank) test = 374.8  on 59 df,   p=<2e-16,   Robust = 3  p=1
## 
##   (Note: the likelihood ratio and score tests assume independence of
##      observations within a cluster, the Wald and robust score tests do not).

92 IPF-Only Quantile-Based G-Computation Multi-Pollutant Survival Analysis - No Bootstrapping

92.1 IPF-Only Simmons Alone

Multi-pollutant model without other covariates aside from dx_yr and site

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr, data=Simm_IPF)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr, data = Simm_IPF)
## 
##           coef exp(coef) se(coef)      z        p
## SO4    0.16647   1.18113  0.21059  0.791 0.429233
## NH4   -0.25333   0.77621  0.68553 -0.370 0.711728
## NO3   -0.01976   0.98043  0.32281 -0.061 0.951185
## BC     0.48917   1.63097  0.52926  0.924 0.355355
## OM     0.22136   1.24777  0.11729  1.887 0.059130
## SS     1.05065   2.85952  0.43253  2.429 0.015137
## Soil  -1.54957   0.21234  0.42420 -3.653 0.000259
## dx_yr  0.04649   1.04759  0.02511  1.851 0.064118
## 
## Likelihood ratio test=36.25  on 8 df, p=1.58e-05
## n= 31982, number of events= 695 
##    (446 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit1 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=Simm_IPF[,c(Xnm, 'start', 'end', 'event', 'dx_yr')], q=4)
qc.survfit1

## Scaled effect size (positive direction, sum of positive coefficients = 0.432)
##    OM   NH4    SS   SO4 
## 0.401 0.241 0.211 0.147 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.226)
##    NO3   Soil     BC 
## 0.5069 0.4529 0.0402 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##      Estimate Std. Error Lower CI Upper CI Z value Pr(>|z|)
## psi1 0.205126   0.096256 0.016467  0.39378   2.131  0.03309

#Lastly the HR is reported through the following
exp(qc.survfit1$coef)

##    psi1 
## 1.22768

exp(qc.survfit1$ci)

## [1] 1.016604 1.482581

Now to plot the findings

plot(qc.survfit1)

Complete multi-pollutant model + dx_yr

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + smokeHx + dich_Race + disadv + dx_yr, data=Simm_IPF)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + 
##     smokeHx + dich_Race + disadv + dx_yr, data = Simm_IPF)
## 
##                         coef exp(coef)  se(coef)      z        p
## SO4                 0.015051  1.015164  0.224522  0.067 0.946555
## NH4                 0.120737  1.128328  0.731716  0.165 0.868940
## NO3                -0.205502  0.814238  0.350773 -0.586 0.557973
## BC                  0.419398  1.521046  0.578392  0.725 0.468384
## OM                  0.295929  1.344375  0.137109  2.158 0.030901
## SS                  0.998684  2.714708  0.469840  2.126 0.033538
## Soil               -1.733707  0.176628  0.450447 -3.849 0.000119
## age_dx              0.001466  1.001467  0.004609  0.318 0.750480
## sexF               -0.366589  0.693094  0.090010 -4.073 4.65e-05
## smokeHxFormer       0.079434  1.082674  0.092474  0.859 0.390352
## smokeHxAlways      -0.586083  0.556503  0.276240 -2.122 0.033868
## smokeHxUnknown      0.396254  1.486246  0.174162  2.275 0.022894
## dich_RaceNon-White  0.132498  1.141677  0.129948  1.020 0.307906
## disadv              0.489431  1.631387  0.138044  3.545 0.000392
## dx_yr               0.028887  1.029309  0.027007  1.070 0.284781
## 
## Likelihood ratio test=80.06  on 15 df, p=6.815e-11
## n= 30488, number of events= 670 
##    (1940 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit2 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=Simm_IPF[,c(Xnm, 'age_dx', 'sex', 'smokeHx', 'dich_Race', 'disadv', 'start', 'end', 'event', 'dx_yr')], q=4)
qc.survfit2

## Scaled effect size (positive direction, sum of positive coefficients = 0.37)
##    OM   NH4    SS   SO4 
## 0.522 0.285 0.160 0.033 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.229)
##    Soil     NO3      BC 
## 0.52211 0.47314 0.00475 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##      Estimate Std. Error  Lower CI Upper CI Z value Pr(>|z|)
## psi1  0.14081    0.10198 -0.059076  0.34069  1.3807   0.1674

#Lastly the HR is reported through the following
exp(qc.survfit2$coef)

##     psi1 
## 1.151204

exp(qc.survfit2$ci)

## [1] 0.9426353 1.4059199

Now to plot the findings

plot(qc.survfit2)

92.2 IPF-Only PFF Alone

Multi-pollutant model without other covariates aside from dx_yr and site

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site, data=PFF_IPF)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site, 
##     data = PFF_IPF)
## 
##              coef exp(coef)  se(coef)      z        p
## SO4      0.086761  1.090636  0.262582  0.330  0.74109
## NH4      0.975197  2.651689  0.613475  1.590  0.11192
## NO3     -0.655077  0.519402  0.222873 -2.939  0.00329
## BC      -0.300052  0.740780  0.480852 -0.624  0.53263
## OM       0.218479  1.244183  0.103502  2.111  0.03478
## SS       0.122207  1.129988  0.157303  0.777  0.43722
## Soil    -0.374684  0.687507  0.266112 -1.408  0.15913
## dx_yr    0.170230  1.185578  0.025657  6.635 3.25e-11
## site02R  0.739631  2.095162  0.395186  1.872  0.06126
## site03R  0.141077  1.151513  0.391011  0.361  0.71825
## site04R  0.207725  1.230874  0.524541  0.396  0.69210
## site05R -0.562068  0.570029  0.447744 -1.255  0.20936
## site06R -0.249953  0.778837  0.393082 -0.636  0.52485
## site07R -0.157797  0.854023  0.374757 -0.421  0.67371
## site08R        NA        NA  0.000000     NA       NA
## site09R -0.131638  0.876658  0.365699 -0.360  0.71888
## site10R  0.139767  1.150006  0.476927  0.293  0.76948
## site11R -0.362254  0.696105  0.381134 -0.950  0.34188
## site12R  0.292830  1.340215  0.379065  0.773  0.43982
## site13R -0.419414  0.657432  0.365631 -1.147  0.25134
## site14R -0.353266  0.702390  0.676924 -0.522  0.60176
## site15R -0.222997  0.800118  0.393831 -0.566  0.57124
## site16R  0.140785  1.151177  0.382385  0.368  0.71274
## site17R  0.119392  1.126812  0.431868  0.276  0.78220
## site18R -0.436953  0.646002  0.395993 -1.103  0.26984
## site19R  0.572791  1.773209  0.495391  1.156  0.24758
## site20R  0.012402  1.012479  0.410612  0.030  0.97591
## site21R -0.519529  0.594800  0.398814 -1.303  0.19268
## site22R -0.048168  0.952974  0.368078 -0.131  0.89588
## site23R -0.039590  0.961184  0.423430 -0.093  0.92551
## site24R  0.277119  1.319323  0.386556  0.717  0.47344
## site25R  0.200238  1.221693  0.384925  0.520  0.60292
## site26R -0.385992  0.679776  0.468053 -0.825  0.40956
## site27R -0.024253  0.976039  0.794356 -0.031  0.97564
## site28R -0.275965  0.758840  0.500139 -0.552  0.58110
## site29R -0.320920  0.725481  0.467091 -0.687  0.49204
## site30R  0.171423  1.186993  0.423976  0.404  0.68598
## site31R -0.513143  0.598611  0.484921 -1.058  0.28996
## site32R  0.383733  1.467754  0.585482  0.655  0.51220
## site33R -0.270238  0.763198  0.458498 -0.589  0.55559
## site34R -0.206830  0.813158  0.353295 -0.585  0.55826
## site35R -0.270013  0.763370  0.350439 -0.770  0.44100
## site36R -0.324174  0.723124  0.408824 -0.793  0.42781
## site37R -0.240132  0.786524  0.404233 -0.594  0.55248
## site38R -0.519216  0.594987  0.385036 -1.348  0.17750
## site39R  0.249465  1.283339  0.392729  0.635  0.52529
## site40R  0.660821  1.936381  0.433942  1.523  0.12780
## site41R -0.149667  0.860994  0.406070 -0.369  0.71244
## site42R  0.002361  1.002364  0.422597  0.006  0.99554
## 
## Likelihood ratio test=130.9  on 48 df, p=1.302e-09
## n= 54499, number of events= 1146 
##    (217 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit1 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=PFF_IPF[,c(Xnm, 'start', 'end', 'event', 'dx_yr', 'site')], q=4)
qc.survfit1

## Scaled effect size (positive direction, sum of positive coefficients = 0.172)
## BC 
##  1 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.321)
##     NO3    Soil     SO4     NH4      OM      SS 
## 0.45420 0.17886 0.16014 0.14257 0.06012 0.00412 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error Lower CI Upper CI Z value Pr(>|z|)
## psi1 -0.148474   0.084657  -0.3144  0.01745 -1.7538  0.07946

#Lastly the HR is reported through the following
exp(qc.survfit1$coef)

##      psi1 
## 0.8620221

exp(qc.survfit1$ci)

## [1] 0.7302278 1.0176031

Now to plot the findings

plot(qc.survfit1)

Complete multi-pollutant model + dx_yr

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + smokeHx + dich_Race + disadv + dx_yr + site, data=PFF_IPF)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + 
##     smokeHx + dich_Race + disadv + dx_yr + site, data = PFF_IPF)
## 
##                          coef  exp(coef)   se(coef)      z        p
## SO4                 0.0682061  1.0705859  0.2643988  0.258  0.79643
## NH4                 1.0312190  2.8044824  0.6192981  1.665  0.09588
## NO3                -0.6928850  0.5001311  0.2271490 -3.050  0.00229
## BC                 -0.3276593  0.7206085  0.4858133 -0.674  0.50002
## OM                  0.2330747  1.2624758  0.1068657  2.181  0.02918
## SS                  0.1335285  1.1428538  0.1588476  0.841  0.40057
## Soil               -0.3176324  0.7278703  0.2716459 -1.169  0.24229
## age_dx              0.0006209  1.0006211  0.0039585  0.157  0.87537
## sexM                0.0668837  1.0691712  0.0714739  0.936  0.34939
## smokeHxEver         0.0432869  1.0442374  0.0646577  0.669  0.50319
## dich_RaceNon-White -0.0784760  0.9245242  0.1266551 -0.620  0.53552
## disadv             -0.0710134  0.9314494  0.1148516 -0.618  0.53637
## dx_yr               0.1686953  1.1837593  0.0262243  6.433 1.25e-10
## site02R             0.6495817  1.9147396  0.4064288  1.598  0.10998
## site03R             0.0477675  1.0489267  0.3999373  0.119  0.90493
## site04R             0.1506611  1.1626026  0.5334735  0.282  0.77763
## site05R            -0.6315396  0.5317725  0.4605219 -1.371  0.17026
## site06R            -0.3148963  0.7298646  0.4099898 -0.768  0.44245
## site07R            -0.2197536  0.8027165  0.3867228 -0.568  0.56987
## site08R                    NA         NA  0.0000000     NA       NA
## site09R            -0.2383156  0.7879540  0.3798596 -0.627  0.53041
## site10R             0.0197230  1.0199187  0.4887279  0.040  0.96781
## site11R            -0.4790718  0.6193580  0.4007408 -1.195  0.23191
## site12R             0.2325779  1.2618487  0.3899961  0.596  0.55094
## site13R            -0.4975654  0.6080091  0.3778694 -1.317  0.18792
## site14R            -0.4295314  0.6508140  0.6831728 -0.629  0.52953
## site15R            -0.3129155  0.7313117  0.4116435 -0.760  0.44716
## site16R             0.0814750  1.0848861  0.3926599  0.207  0.83562
## site17R             0.0361574  1.0368191  0.4446305  0.081  0.93519
## site18R            -0.5295561  0.5888663  0.4115308 -1.287  0.19817
## site19R             0.4301527  1.5374923  0.5035536  0.854  0.39298
## site20R            -0.0708739  0.9315794  0.4218986 -0.168  0.86659
## site21R            -0.6095632  0.5435883  0.4141382 -1.472  0.14105
## site22R            -0.1109382  0.8949941  0.3806988 -0.291  0.77074
## site23R            -0.1693367  0.8442246  0.4362460 -0.388  0.69789
## site24R             0.2147689  1.2395755  0.3969884  0.541  0.58851
## site25R             0.1244613  1.1325381  0.3932679  0.316  0.75164
## site26R            -0.4979039  0.6078034  0.4790399 -1.039  0.29863
## site27R            -0.1303783  0.8777633  0.8044672 -0.162  0.87125
## site28R            -0.3239281  0.7233022  0.5129817 -0.631  0.52774
## site29R            -0.3988858  0.6710674  0.4769630 -0.836  0.40298
## site30R             0.1227835  1.1306396  0.4376044  0.281  0.77903
## site31R            -0.5914321  0.5535340  0.4944519 -1.196  0.23164
## site32R             0.1943749  1.2145516  0.6068172  0.320  0.74873
## site33R            -0.3820941  0.6824308  0.4703026 -0.812  0.41654
## site34R            -0.2718159  0.7619945  0.3672039 -0.740  0.45916
## site35R            -0.3593655  0.6981191  0.3656186 -0.983  0.32566
## site36R            -0.3796064  0.6841306  0.4237750 -0.896  0.37037
## site37R            -0.2916583  0.7470238  0.4161383 -0.701  0.48338
## site38R            -0.5818764  0.5588488  0.3995920 -1.456  0.14534
## site39R             0.1481103  1.1596408  0.4014249  0.369  0.71216
## site40R             0.5907592  1.8053585  0.4440327  1.330  0.18337
## site41R            -0.2109789  0.8097911  0.4162127 -0.507  0.61222
## site42R            -0.0768702  0.9260100  0.4335103 -0.177  0.85926
## 
## Likelihood ratio test=131.6  on 53 df, p=1.256e-08
## n= 53867, number of events= 1131 
##    (849 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit2 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=PFF_IPF[,c(Xnm, 'age_dx', 'sex', 'smokeHx', 'dich_Race', 'disadv', 'start', 'end', 'event', 'dx_yr', 'site')], q=4)
qc.survfit2

## Scaled effect size (positive direction, sum of positive coefficients = 0.176)
## BC 
##  1 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.329)
##     NO3     NH4    Soil     SO4      OM      SS 
## 0.44361 0.18175 0.17572 0.15412 0.03593 0.00886 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error Lower CI Upper CI Z value Pr(>|z|)
## psi1 -0.152718   0.085593 -0.32048 0.015041 -1.7842  0.07439

#Lastly the HR is reported through the following
exp(qc.survfit2$coef)

##      psi1 
## 0.8583717

exp(qc.survfit2$ci)

## [1] 0.7258023 1.0151551

Now to plot the findings

plot(qc.survfit2)

92.3 IPF-Only CARE Alone

Multi-pollutant model without other covariates aside from dx_yr and site

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site, data=CARE_IPF)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site, 
##     data = CARE_IPF)
## 
##             coef exp(coef) se(coef)      z        p
## SO4     -0.78591   0.45570  0.69324 -1.134  0.25693
## NH4     -4.12997   0.01608  1.40466 -2.940  0.00328
## NO3      2.04128   7.70048  0.45594  4.477 7.57e-06
## BC       3.21733  24.96134  1.10532  2.911  0.00361
## OM      -0.73315   0.48040  0.15687 -4.674 2.96e-06
## SS      -0.46123   0.63050  0.53764 -0.858  0.39096
## Soil     0.99067   2.69303  0.53018  1.869  0.06169
## dx_yr    0.61189   1.84391  0.03943 15.518  < 2e-16
## site102 -0.24412   0.78339  0.28595 -0.854  0.39326
## site103  0.32255   1.38064  0.31749  1.016  0.30966
## site104  0.19787   1.21880  0.32606  0.607  0.54396
## site105  0.61776   1.85476  0.19018  3.248  0.00116
## site106  0.22905   1.25741  0.13279  1.725  0.08455
## site107  0.73374   2.08286  0.24637  2.978  0.00290
## site108 -0.99633   0.36923  0.32183 -3.096  0.00196
## 
## Likelihood ratio test=1125  on 15 df, p=< 2.2e-16
## n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit1 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=CARE_IPF[,c(Xnm, 'start', 'end', 'event', 'dx_yr', 'site')], q=4)
qc.survfit1

## Scaled effect size (positive direction, sum of positive coefficients = 0.261)
##     BC    NO3   Soil 
## 0.5490 0.3611 0.0898 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.509)
##     OM    NH4    SO4     SS 
## 0.3764 0.3693 0.1956 0.0587 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error Lower CI  Upper CI Z value Pr(>|z|)
## psi1 -0.248189   0.086547 -0.41782 -0.078561 -2.8677 0.004135

#Lastly the HR is reported through the following
exp(qc.survfit1$coef)

##      psi1 
## 0.7802122

exp(qc.survfit1$ci)

## [1] 0.6584821 0.9244459

Now to plot the findings

plot(qc.survfit1)

Complete multi-pollutant model + dx_yr

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + smokeHx + dich_Race + disadv + dx_yr + site, data=CARE_IPF)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + 
##     smokeHx + dich_Race + disadv + dx_yr + site, data = CARE_IPF)
## 
##                         coef exp(coef)  se(coef)      z        p
## SO4                -0.646700  0.523771  0.709768 -0.911  0.36222
## NH4                -4.348431  0.012927  1.426703 -3.048  0.00230
## NO3                 2.075928  7.971944  0.463491  4.479 7.50e-06
## BC                  3.211722 24.821783  1.118638  2.871  0.00409
## OM                 -0.737455  0.478330  0.158416 -4.655 3.24e-06
## SS                 -0.496937  0.608392  0.543497 -0.914  0.36054
## Soil                0.878388  2.407016  0.540870  1.624  0.10437
## age_dx              0.012854  1.012937  0.004582  2.805  0.00503
## sexF               -0.036144  0.964502  0.077360 -0.467  0.64035
## smokeHxFormer      -0.006665  0.993358  0.081933 -0.081  0.93517
## smokeHxAlways       0.042406  1.043318  0.170431  0.249  0.80350
## smokeHxUnknown      0.560480  1.751513  0.737086  0.760  0.44702
## dich_RaceNon-White -0.108693  0.897005  0.111837 -0.972  0.33111
## disadv             -0.019298  0.980887  0.132027 -0.146  0.88379
## dx_yr               0.612491  1.845022  0.039797 15.390  < 2e-16
## site102            -0.231676  0.793203  0.293406 -0.790  0.42976
## site103             0.357804  1.430186  0.328091  1.091  0.27546
## site104             0.234442  1.264203  0.334763  0.700  0.48373
## site105             0.640986  1.898352  0.195877  3.272  0.00107
## site106             0.202590  1.224571  0.134152  1.510  0.13100
## site107             0.672944  1.960000  0.255484  2.634  0.00844
## site108            -1.015570  0.362196  0.330616 -3.072  0.00213
## 
## Likelihood ratio test=1136  on 22 df, p=< 2.2e-16
## n= 36886, number of events= 908 
##    (60 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit2 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=CARE_IPF[,c(Xnm, 'age_dx', 'sex', 'smokeHx', 'dich_Race', 'disadv', 'start', 'end', 'event', 'dx_yr', 'site')], q=4)
qc.survfit2

## Scaled effect size (positive direction, sum of positive coefficients = 0.254)
##     BC    NO3   Soil 
## 0.5598 0.3960 0.0442 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.511)
##    NH4     OM    SO4     SS 
## 0.3793 0.3752 0.1834 0.0621 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error Lower CI  Upper CI Z value Pr(>|z|)
## psi1 -0.257038   0.087759 -0.42904 -0.085034 -2.9289 0.003402

#Lastly the HR is reported through the following
exp(qc.survfit2$coef)

##      psi1 
## 0.7733387

exp(qc.survfit2$ci)

## [1] 0.6511322 0.9184813

Now to plot the findings

plot(qc.survfit2)

92.4 IPF-Only One-Stage Meta-Analysis - i.e. Combined Cohorts

Multi-pollutant model without other covariates aside from dx_yr and site

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site + cluster(cohort), data=All_IPF)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + dx_yr + site, 
##     data = All_IPF, cluster = cohort)
## 
##             coef exp(coef) se(coef) robust se       z        p
## SO4      0.73035   2.07581  0.12006   0.40766   1.792   0.0732
## NH4     -0.55779   0.57247  0.36695   1.04949  -0.531   0.5951
## NO3     -0.12097   0.88606  0.15794   0.42204  -0.287   0.7744
## BC       1.05796   2.88050  0.27318   2.05141   0.516   0.6060
## OM      -0.17272   0.84137  0.05788   0.39206  -0.441   0.6595
## SS       0.11210   1.11863  0.14063   0.24561   0.456   0.6481
## Soil    -0.35240   0.70300  0.19642   0.20120  -1.752   0.0799
## dx_yr    0.21263   1.23693  0.01284   0.09938   2.140   0.0324
## site02R  0.41126   1.50872  0.38497   0.40873   1.006   0.3143
## site03R -0.16364   0.84905  0.37195   0.34792  -0.470   0.6381
## site04R -0.78060   0.45813  0.48445   0.72554  -1.076   0.2820
## site05R -1.28198   0.27749  0.38712   0.62990  -2.035   0.0418
## site06R -0.87299   0.41770  0.33957   0.53135  -1.643   0.1004
## site07R -0.83772   0.43270  0.33125   0.49371  -1.697   0.0897
## site09R -0.59399   0.55212  0.34646   0.33842  -1.755   0.0792
## site1   -0.96789   0.37988  0.30370   0.72637  -1.332   0.1827
## site101 -0.57829   0.56086  0.30339   0.34687  -1.667   0.0955
## site102 -0.19020   0.82680  0.30128   0.03508  -5.422 5.91e-08
## site103 -0.02096   0.97926  0.28708   0.01706  -1.229   0.2191
## site104 -0.35820   0.69893  0.29733   0.01870 -19.158  < 2e-16
## site105 -0.28071   0.75525  0.29499   0.05868  -4.784 1.72e-06
## site106 -0.71282   0.49026  0.29997   0.35878  -1.987   0.0469
## site107 -0.19773   0.82059  0.33171   0.07768  -2.545   0.0109
## site108 -0.09243   0.91172  0.37631   0.24443  -0.378   0.7053
## site10R -0.61826   0.53888  0.42835   0.49308  -1.254   0.2099
## site11R -0.93599   0.39220  0.32084   0.61491  -1.522   0.1280
## site12R -0.24707   0.78109  0.35472   0.32909  -0.751   0.4528
## site13R -1.18214   0.30662  0.31897   0.67134  -1.761   0.0783
## site14R -1.11988   0.32632  0.65283   0.57665  -1.942   0.0521
## site15R -0.84908   0.42781  0.34455   0.53351  -1.591   0.1115
## site16R -0.49679   0.60848  0.35729   0.47612  -1.043   0.2968
## site17R -0.60207   0.54768  0.38493   0.56101  -1.073   0.2832
## site18R -1.02728   0.35798  0.34154   0.55637  -1.846   0.0648
## site19R  0.19359   1.21360  0.43400   0.45682   0.424   0.6717
## site20R -0.65259   0.52069  0.37154   0.47851  -1.364   0.1726
## site21R -1.29100   0.27500  0.34304   0.79401  -1.626   0.1040
## site22R -0.86410   0.42143  0.32873   0.67724  -1.276   0.2020
## site23R -0.62165   0.53706  0.36325   0.50428  -1.233   0.2177
## site24R -0.30381   0.73800  0.35563   0.34147  -0.890   0.3736
## site25R -0.18337   0.83246  0.37300   0.38861  -0.472   0.6370
## site26R -1.05685   0.34755  0.39346   0.56193  -1.881   0.0600
## site27R -0.89743   0.40762  0.76996   0.70017  -1.282   0.1999
## site28R -1.06677   0.34412  0.45256   0.66379  -1.607   0.1080
## site29R -1.01452   0.36258  0.44411   0.48787  -2.079   0.0376
## site30R -0.85189   0.42661  0.37761   0.83729  -1.017   0.3089
## site31R -1.13490   0.32145  0.45654   0.57285  -1.981   0.0476
## site32R -0.01600   0.98413  0.50069   0.61675  -0.026   0.9793
## site33R -0.95322   0.38550  0.38280   0.61682  -1.545   0.1223
## site34R -0.73068   0.48158  0.32282   0.38565  -1.895   0.0581
## site35R -0.59488   0.55163  0.33711   0.23034  -2.583   0.0098
## site36R -0.99933   0.36813  0.35892   0.56125  -1.781   0.0750
## site37R -1.01398   0.36277  0.35760   0.63391  -1.600   0.1097
## site38R -1.13204   0.32237  0.33945   0.51400  -2.202   0.0276
## site39R -0.05302   0.94836  0.38023   0.30602  -0.173   0.8625
## site40R -0.17189   0.84207  0.40685   1.00296  -0.171   0.8639
## site41R -0.87545   0.41668  0.38396   0.64137  -1.365   0.1723
## site42R -0.80581   0.44672  0.37843   0.64566  -1.248   0.2120
## 
## Likelihood ratio test=528.8  on 57 df, p=< 2.2e-16
## n= 123367, number of events= 2749 
##    (723 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit1 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=All_IPF[,c(Xnm, 'start', 'end', 'event', 'dx_yr', 'site', 'cohort')], q=4)
qc.survfit1

## Scaled effect size (positive direction, sum of positive coefficients = 0.31)
##     BC   Soil    SO4 
## 0.7345 0.2260 0.0396 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.365)
##     NH4      OM     NO3      SS 
## 0.53449 0.35055 0.10544 0.00952 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error Lower CI Upper CI Z value Pr(>|z|)
## psi1 -0.054871   0.053901 -0.16051 0.050773  -1.018   0.3087

#Lastly the HR is reported through the following
exp(qc.survfit1$coef)

##      psi1 
## 0.9466076

exp(qc.survfit1$ci)

## [1] 0.8517055 1.0520843

Now to plot the findings

plot(qc.survfit1)

Complete multi-pollutant model

#First need to list the pollutants in the model
Xnm <- c('SO4','NO3','NH4','BC','OM','SS','Soil')

#Next construct the base cox model
survival::coxph(survival::Surv(start, end, event==1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + smokeHx + dich_Race + disadv + dx_yr + site + cluster(cohort), data=All_IPF)

## Call:
## survival::coxph(formula = survival::Surv(start, end, event == 
##     1) ~ SO4 + NH4 + NO3 + BC + OM + SS + Soil + age_dx + sex + 
##     smokeHx + dich_Race + disadv + dx_yr + site, data = All_IPF, 
##     cluster = cohort)
## 
##                         coef exp(coef)  se(coef) robust se       z        p
## SO4                 0.729102  2.073218  0.122054  0.413807   1.762 0.078080
## NH4                -0.526217  0.590836  0.371985  1.081748  -0.486 0.626648
## NO3                -0.116989  0.889595  0.160290  0.446826  -0.262 0.793458
## BC                  1.115783  3.051957  0.277513  2.089164   0.534 0.593286
## OM                 -0.191679  0.825571  0.058941  0.391135  -0.490 0.624092
## SS                  0.092221  1.096608  0.142145  0.236987   0.389 0.697171
## Soil               -0.313375  0.730976  0.199709  0.199586  -1.570 0.116385
## age_dx              0.002221  1.002224  0.002437  0.001999   1.111 0.266453
## sexF               -0.150308  0.860443  0.044975  0.084643  -1.776 0.075769
## smokeHxFormer       0.114838  1.121692  0.059665  0.016400   7.002 2.52e-12
## smokeHxAlways      -0.004764  0.995248  0.138843  0.106168  -0.045 0.964211
## smokeHxUnknown      0.323062  1.381351  0.154155  0.179538   1.799 0.071954
## smokeHxEver         0.031876  1.032390  0.064116  0.006932   4.598 4.26e-06
## dich_RaceNon-White  0.026473  1.026826  0.069198  0.086931   0.305 0.760727
## disadv              0.089470  1.093594  0.072029  0.103013   0.869 0.385108
## dx_yr               0.212487  1.236749  0.013249  0.104647   2.030 0.042306
## site02R             0.403773  1.497464  0.394718  0.354074   1.140 0.254135
## site03R            -0.190536  0.826516  0.380949  0.357920  -0.532 0.594489
## site04R            -0.829753  0.436157  0.492570  0.747281  -1.110 0.266842
## site05R            -1.330807  0.264264  0.397853  0.646559  -2.058 0.039562
## site06R            -0.942449  0.389672  0.352865  0.528065  -1.785 0.074307
## site07R            -0.896980  0.407799  0.342156  0.502405  -1.785 0.074201
## site09R            -0.659495  0.517112  0.358616  0.338902  -1.946 0.051658
## site1              -1.106646  0.330666  0.320806  0.719791  -1.537 0.124182
## site101            -0.679735  0.506751  0.318362  0.351108  -1.936 0.052871
## site102            -0.287600  0.750062  0.316691  0.040356  -7.127 1.03e-12
## site103            -0.123630  0.883707  0.303150  0.032266  -3.832 0.000127
## site104            -0.449816  0.637745  0.313276  0.029836 -15.077  < 2e-16
## site105            -0.379158  0.684438  0.312583  0.058970  -6.430 1.28e-10
## site106            -0.842435  0.430660  0.315643  0.363728  -2.316 0.020552
## site107            -0.320169  0.726026  0.347582  0.081596  -3.924 8.71e-05
## site108            -0.213308  0.807907  0.388678  0.247231  -0.863 0.388255
## site10R            -0.715393  0.489000  0.439085  0.460421  -1.554 0.120237
## site11R            -1.009486  0.364406  0.335433  0.621776  -1.624 0.104472
## site12R            -0.317514  0.727956  0.365021  0.341618  -0.929 0.352660
## site13R            -1.253239  0.285578  0.330622  0.680305  -1.842 0.065450
## site14R            -1.162674  0.312649  0.658714  0.597193  -1.947 0.051547
## site15R            -0.918264  0.399211  0.357475  0.524733  -1.750 0.080125
## site16R            -0.538755  0.583474  0.367666  0.473905  -1.137 0.255604
## site17R            -0.679300  0.506972  0.395713  0.573417  -1.185 0.236155
## site18R            -1.102539  0.332027  0.353195  0.542070  -2.034 0.041957
## site19R             0.083009  1.086552  0.443305  0.441326   0.188 0.850806
## site20R            -0.703426  0.494887  0.382051  0.483568  -1.455 0.145764
## site21R            -1.365772  0.255184  0.355168  0.800215  -1.707 0.087867
## site22R            -0.919568  0.398691  0.340847  0.696206  -1.321 0.186559
## site23R            -0.736183  0.478939  0.375571  0.488699  -1.506 0.131961
## site24R            -0.356308  0.700257  0.366199  0.346409  -1.029 0.303678
## site25R            -0.195842  0.822142  0.381138  0.330353  -0.593 0.553297
## site26R            -1.156147  0.314697  0.402971  0.532325  -2.172 0.029865
## site27R            -0.939588  0.390789  0.777294  0.685246  -1.371 0.170322
## site28R            -1.124445  0.324833  0.462113  0.682983  -1.646 0.099687
## site29R            -1.033247  0.355849  0.453196  0.504043  -2.050 0.040372
## site30R            -0.968753  0.379556  0.390676  0.871411  -1.112 0.266264
## site31R            -1.201218  0.300828  0.464993  0.575538  -2.087 0.036877
## site32R            -0.168618  0.844831  0.519749  0.624164  -0.270 0.787044
## site33R            -1.047076  0.350963  0.393788  0.585604  -1.788 0.073772
## site34R            -0.774278  0.461037  0.335019  0.396589  -1.952 0.050898
## site35R            -0.655435  0.519216  0.350174  0.225141  -2.911 0.003600
## site36R            -1.059422  0.346656  0.370409  0.585160  -1.810 0.070221
## site37R            -1.078461  0.340118  0.368640  0.660571  -1.633 0.102549
## site38R            -1.177667  0.307996  0.351464  0.504588  -2.334 0.019600
## site39R            -0.134541  0.874117  0.390189  0.315008  -0.427 0.669303
## site40R            -0.288640  0.749282  0.417646  1.030048  -0.280 0.779309
## site41R            -0.960294  0.382780  0.393877  0.656131  -1.464 0.143311
## site42R            -0.869892  0.418997  0.388589  0.651225  -1.336 0.181622
## 
## Likelihood ratio test=561.7  on 65 df, p=< 2.2e-16
## n= 121241, number of events= 2709 
##    (2849 observations deleted due to missingness)

#Next the quantile regression model
qc.survfit2 <- qgcomp.cox.noboot(survival::Surv(start, end, event==1) ~ ., expnms=Xnm, data=All_IPF[,c(Xnm, 'age_dx', 'sex', 'smokeHx', 'dich_Race', 'disadv', 'start', 'end', 'event', 'dx_yr', 'site', 'cohort')], q=4)
qc.survfit2

## Scaled effect size (positive direction, sum of positive coefficients = 0.311)
##     BC   Soil    SO4 
## 0.7234 0.2278 0.0488 
## 
## Scaled effect size (negative direction, sum of negative coefficients = -0.371)
##    NH4     OM    NO3     SS 
## 0.5239 0.3485 0.1031 0.0245 
## 
## Mixture log(hazard ratio) (Delta method CI):
## 
##       Estimate Std. Error Lower CI Upper CI Z value Pr(>|z|)
## psi1 -0.060520   0.054604 -0.16754 0.046501 -1.1084   0.2677

#Lastly the HR is reported through the following
exp(qc.survfit2$coef)

##     psi1 
## 0.941275

exp(qc.survfit2$ci)

## [1] 0.845742 1.047599

Now to plot the findings

plot(qc.survfit2)